Merge branch 'main' of github.com:pytorch/vision into ufmt

Author: NicolasHug

gallery/plot_visualization_utils.py

@@ -116,7 +116,7 @@ def show(imgs):
 # lraspp mobilenet models
 # (:func:`~torchvision.models.segmentation.lraspp_mobilenet_v3_large`).
 #
-# Let's start by looking at the ouput of the model. Remember that in general,
+# Let's start by looking at the output of the model. Remember that in general,
 # images must be normalized before they're passed to a semantic segmentation
 # model.
 

torchvision/models/detection/rpn.py

@@ -230,7 +230,7 @@ def filter_proposals(self, proposals, objectness, image_shapes, num_anchors_per_
         # type: (Tensor, Tensor, List[Tuple[int, int]], List[int]) -> Tuple[List[Tensor], List[Tensor]]
         num_images = proposals.shape[0]
         device = proposals.device
-        # do not backprop throught objectness
+        # do not backprop through objectness
         objectness = objectness.detach()
         objectness = objectness.reshape(num_images, -1)
 

torchvision/models/feature_extraction.py

@@ -37,7 +37,7 @@ class NodePathTracer(LeafModuleAwareTracer):
     """
     NodePathTracer is an FX tracer that, for each operation, also records the
     name of the Node from which the operation originated. A node name here is
-    a `.` seperated path walking the hierarchy from top level module down to
+    a `.` separated path walking the hierarchy from top level module down to
     leaf operation or leaf module. The name of the top level module is not
     included as part of the node name. For example, if we trace a module whose
     forward method applies a ReLU module, the name for that node will simply

torchvision/ops/ps_roi_align.py

@@ -30,8 +30,10 @@ def ps_roi_align(
             in the batch.
         output_size (int or Tuple[int, int]): the size of the output (in bins or pixels) after the pooling
             is performed, as (height, width).
-        spatial_scale (float): a scaling factor that maps the input coordinates to
-            the box coordinates. Default: 1.0
+        spatial_scale (float): a scaling factor that maps the box coordinates to
+            the input coordinates. For example, if your boxes are defined on the scale
+            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+            the original image), you'll want to set this to 0.5. Default: 1.0
         sampling_ratio (int): number of sampling points in the interpolation grid
             used to compute the output value of each pooled output bin. If > 0,
             then exactly ``sampling_ratio x sampling_ratio`` sampling points per bin are used. If

torchvision/ops/ps_roi_pool.py

@@ -29,8 +29,10 @@ def ps_roi_pool(
             in the batch.
         output_size (int or Tuple[int, int]): the size of the output (in bins or pixels) after the pooling
             is performed, as (height, width).
-        spatial_scale (float): a scaling factor that maps the input coordinates to
-            the box coordinates. Default: 1.0
+        spatial_scale (float): a scaling factor that maps the box coordinates to
+            the input coordinates. For example, if your boxes are defined on the scale
+            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+            the original image), you'll want to set this to 0.5. Default: 1.0
 
     Returns:
         Tensor[K, C / (output_size[0] * output_size[1]), output_size[0], output_size[1]]: The pooled RoIs.

torchvision/ops/roi_align.py

@@ -33,8 +33,10 @@ def roi_align(
             in the batch.
         output_size (int or Tuple[int, int]): the size of the output (in bins or pixels) after the pooling
             is performed, as (height, width).
-        spatial_scale (float): a scaling factor that maps the input coordinates to
-            the box coordinates. Default: 1.0
+        spatial_scale (float): a scaling factor that maps the box coordinates to
+            the input coordinates. For example, if your boxes are defined on the scale
+            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+            the original image), you'll want to set this to 0.5. Default: 1.0
         sampling_ratio (int): number of sampling points in the interpolation grid
             used to compute the output value of each pooled output bin. If > 0,
             then exactly ``sampling_ratio x sampling_ratio`` sampling points per bin are used. If

torchvision/ops/roi_pool.py

@@ -30,8 +30,10 @@ def roi_pool(
             in the batch.
         output_size (int or Tuple[int, int]): the size of the output after the cropping
             is performed, as (height, width)
-        spatial_scale (float): a scaling factor that maps the input coordinates to
-            the box coordinates. Default: 1.0
+        spatial_scale (float): a scaling factor that maps the box coordinates to
+            the input coordinates. For example, if your boxes are defined on the scale
+            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+            the original image), you'll want to set this to 0.5. Default: 1.0
 
     Returns:
         Tensor[K, C, output_size[0], output_size[1]]: The pooled RoIs.

torchvision/prototype/datasets/_builtin/caltech.py

@@ -120,8 +120,7 @@ def _make_datapipe(
 
         images_dp = TarArchiveReader(images_dp)
         images_dp = Filter(images_dp, self._is_not_background_image)
-        # FIXME: add this after https://github.com/pytorch/pytorch/issues/65808 is resolved
-        # images_dp = Shuffler(images_dp, buffer_size=INFINITE_BUFFER_SIZE)
+        images_dp = Shuffler(images_dp, buffer_size=INFINITE_BUFFER_SIZE)
 
         anns_dp = TarArchiveReader(anns_dp)
         anns_dp = Filter(anns_dp, self._is_ann)
@@ -189,8 +188,7 @@ def _make_datapipe(
         dp = resource_dps[0]
         dp = TarArchiveReader(dp)
         dp = Filter(dp, self._is_not_rogue_file)
-        # FIXME: add this after https://github.com/pytorch/pytorch/issues/65808 is resolved
-        # dp = Shuffler(dp, buffer_size=INFINITE_BUFFER_SIZE)
+        dp = Shuffler(dp, buffer_size=INFINITE_BUFFER_SIZE)
         return Mapper(dp, self._collate_and_decode_sample, fn_kwargs=dict(decoder=decoder))
 
     def generate_categories_file(self, root: Union[str, pathlib.Path]) -> None: