Adding multiweight support to Quantized ResNet (#4827)

* Adding multi-weight support to Quantized ResNet.

* Update references script to support testing quantized models with the new API.

* Handle quantized models correctly in ref script.

* Fixing references for quantization.

Author: datumbox

references/classification/README.md

@@ -151,9 +151,9 @@ torchrun --nproc_per_node=8 train.py\
 
 ## Quantized
 
-### Parameters used for generating quantized models:
+### Post training quantized models
 
-For all post training quantized models (All quantized models except mobilenet-v2), the settings are:
+For all post training quantized models, the settings are:
 
 1. num_calibration_batches: 32
 2. num_workers: 16
@@ -162,8 +162,11 @@ For all post training quantized models (All quantized models except mobilenet-v2
 5. backend: 'fbgemm'
 
 ```
-python train_quantization.py --device='cpu' --post-training-quantize --backend='fbgemm' --model='<model_name>'
+python train_quantization.py --device='cpu' --post-training-quantize --backend='fbgemm' --model='$MODEL'
 ```
+Here `$MODEL` is one of `googlenet`, `inception_v3`, `resnet18`, `resnet50`, `resnext101_32x8d` and `shufflenet_v2_x1_0`.
+
+### QAT MobileNetV2
 
 For Mobilenet-v2, the model was trained with quantization aware training, the settings used are:
 1. num_workers: 16
@@ -185,6 +188,8 @@ torchrun --nproc_per_node=8 train_quantization.py --model='mobilenet_v2'
 
 Training converges at about 10 epochs.
 
+### QAT MobileNetV3
+
 For Mobilenet-v3 Large, the model was trained with quantization aware training, the settings used are:
 1. num_workers: 16
 2. batch_size: 32

references/classification/train.py

@@ -153,7 +153,7 @@ def load_data(traindir, valdir, args):
                 crop_size=val_crop_size, resize_size=val_resize_size, interpolation=interpolation
             )
         else:
-            fn = PM.__dict__[args.model]
+            fn = PM.quantization.__dict__[args.model] if hasattr(args, "backend") else PM.__dict__[args.model]
             weights = PM._api.get_weight(fn, args.weights)
             preprocessing = weights.transforms()
 

references/classification/train_quantization.py

@@ -12,6 +12,12 @@
 from train import train_one_epoch, evaluate, load_data
 
 
+try:
+    from torchvision.prototype import models as PM
+except ImportError:
+    PM = None
+
+
 def main(args):
     if args.output_dir:
         utils.mkdir(args.output_dir)
@@ -46,7 +52,12 @@ def main(args):
 
     print("Creating model", args.model)
     # when training quantized models, we always start from a pre-trained fp32 reference model
-    model = torchvision.models.quantization.__dict__[args.model](pretrained=True, quantize=args.test_only)
+    if not args.weights:
+        model = torchvision.models.quantization.__dict__[args.model](pretrained=True, quantize=args.test_only)
+    else:
+        if PM is None:
+            raise ImportError("The prototype module couldn't be found. Please install the latest torchvision nightly.")
+        model = PM.quantization.__dict__[args.model](weights=args.weights, quantize=args.test_only)
     model.to(device)
 
     if not (args.test_only or args.post_training_quantize):
@@ -251,6 +262,9 @@ def get_args_parser(add_help=True):
         "--train-crop-size", default=224, type=int, help="the random crop size used for training (default: 224)"
     )
 
+    # Prototype models only
+    parser.add_argument("--weights", default=None, type=str, help="the weights enum name to load")
+
     return parser
 
 

test/test_prototype_models.py

@@ -30,10 +30,15 @@ def get_models_with_module_names(module):
     return [(fn, module_name) for fn in TM.get_models_from_module(module)]
 
 
-def test_get_weight():
-    fn = models.resnet50
-    weight_name = "ImageNet1K_RefV2"
-    assert models._api.get_weight(fn, weight_name) == models.ResNet50Weights.ImageNet1K_RefV2
+@pytest.mark.parametrize(
+    "model_fn, weight",
+    [
+        (models.resnet50, models.ResNet50Weights.ImageNet1K_RefV2),
+        (models.quantization.resnet50, models.quantization.QuantizedResNet50Weights.ImageNet1K_FBGEMM_RefV1),
+    ],
+)
+def test_get_weight(model_fn, weight):
+    assert models._api.get_weight(model_fn, weight.name) == weight
 
 
 @pytest.mark.parametrize("model_fn", TM.get_models_from_module(models))
@@ -43,6 +48,12 @@ def test_classification_model(model_fn, dev):
     TM.test_classification_model(model_fn, dev)
 
 
+@pytest.mark.parametrize("model_fn", TM.get_models_from_module(models.quantization))
+@pytest.mark.skipif(os.getenv("PYTORCH_TEST_WITH_PROTOTYPE", "0") == "0", reason="Prototype code tests are disabled")
+def test_quantized_classification_model(model_fn):
+    TM.test_quantized_classification_model(model_fn)
+
+
 @pytest.mark.parametrize("model_fn", TM.get_models_from_module(models.segmentation))
 @pytest.mark.parametrize("dev", cpu_and_gpu())
 @pytest.mark.skipif(os.getenv("PYTORCH_TEST_WITH_PROTOTYPE", "0") == "0", reason="Prototype code tests are disabled")
@@ -60,6 +71,7 @@ def test_video_model(model_fn, dev):
 @pytest.mark.parametrize(
     "model_fn, module_name",
     get_models_with_module_names(models)
+    + get_models_with_module_names(models.quantization)
     + get_models_with_module_names(models.segmentation)
     + get_models_with_module_names(models.video),
 )
@@ -70,6 +82,9 @@ def test_old_vs_new_factory(model_fn, module_name, dev):
         "models": {
             "input_shape": (1, 3, 224, 224),
         },
+        "quantization": {
+            "input_shape": (1, 3, 224, 224),
+        },
         "segmentation": {
             "input_shape": (1, 3, 520, 520),
         },

torchvision/prototype/models/quantization/resnet.py

@@ -12,10 +12,18 @@
 from ...transforms.presets import ImageNetEval
 from .._api import Weights, WeightEntry
 from .._meta import _IMAGENET_CATEGORIES
-from ..resnet import ResNet50Weights
+from ..resnet import ResNet18Weights, ResNet50Weights, ResNeXt101_32x8dWeights
 
 
-__all__ = ["QuantizableResNet", "QuantizedResNet50Weights", "resnet50"]
+__all__ = [
+    "QuantizableResNet",
+    "QuantizedResNet18Weights",
+    "QuantizedResNet50Weights",
+    "QuantizedResNeXt101_32x8dWeights",
+    "resnet18",
+    "resnet50",
+    "resnext101_32x8d",
+]
 
 
 def _resnet(
@@ -47,22 +55,67 @@ def _resnet(
     "size": (224, 224),
     "categories": _IMAGENET_CATEGORIES,
     "backend": "fbgemm",
+    "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models",
 }
 
 
+class QuantizedResNet18Weights(Weights):
+    ImageNet1K_FBGEMM_RefV1 = WeightEntry(
+        url="https://download.pytorch.org/models/quantized/resnet18_fbgemm_16fa66dd.pth",
+        transforms=partial(ImageNetEval, crop_size=224),
+        meta={
+            **_common_meta,
+            "acc@1": 69.494,
+            "acc@5": 88.882,
+        },
+    )
+
+
 class QuantizedResNet50Weights(Weights):
     ImageNet1K_FBGEMM_RefV1 = WeightEntry(
         url="https://download.pytorch.org/models/quantized/resnet50_fbgemm_bf931d71.pth",
         transforms=partial(ImageNetEval, crop_size=224),
         meta={
             **_common_meta,
-            "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#quantized",
             "acc@1": 75.920,
             "acc@5": 92.814,
         },
     )
 
 
+class QuantizedResNeXt101_32x8dWeights(Weights):
+    ImageNet1K_FBGEMM_RefV1 = WeightEntry(
+        url="https://download.pytorch.org/models/quantized/resnext101_32x8_fbgemm_09835ccf.pth",
+        transforms=partial(ImageNetEval, crop_size=224),
+        meta={
+            **_common_meta,
+            "acc@1": 78.986,
+            "acc@5": 94.480,
+        },
+    )
+
+
+def resnet18(
+    weights: Optional[Union[QuantizedResNet18Weights, ResNet18Weights]] = None,
+    progress: bool = True,
+    quantize: bool = False,
+    **kwargs: Any,
+) -> QuantizableResNet:
+    if "pretrained" in kwargs:
+        warnings.warn("The argument pretrained is deprecated, please use weights instead.")
+        if kwargs.pop("pretrained"):
+            weights = QuantizedResNet18Weights.ImageNet1K_FBGEMM_RefV1 if quantize else ResNet18Weights.ImageNet1K_RefV1
+        else:
+            weights = None
+
+    if quantize:
+        weights = QuantizedResNet18Weights.verify(weights)
+    else:
+        weights = ResNet18Weights.verify(weights)
+
+    return _resnet(QuantizableBasicBlock, [2, 2, 2, 2], weights, progress, quantize, **kwargs)
+
+
 def resnet50(
     weights: Optional[Union[QuantizedResNet50Weights, ResNet50Weights]] = None,
     progress: bool = True,
@@ -82,3 +135,30 @@ def resnet50(
         weights = ResNet50Weights.verify(weights)
 
     return _resnet(QuantizableBottleneck, [3, 4, 6, 3], weights, progress, quantize, **kwargs)
+
+
+def resnext101_32x8d(
+    weights: Optional[Union[QuantizedResNeXt101_32x8dWeights, ResNeXt101_32x8dWeights]] = None,
+    progress: bool = True,
+    quantize: bool = False,
+    **kwargs: Any,
+) -> QuantizableResNet:
+    if "pretrained" in kwargs:
+        warnings.warn("The argument pretrained is deprecated, please use weights instead.")
+        if kwargs.pop("pretrained"):
+            weights = (
+                QuantizedResNeXt101_32x8dWeights.ImageNet1K_FBGEMM_RefV1
+                if quantize
+                else ResNeXt101_32x8dWeights.ImageNet1K_RefV1
+            )
+        else:
+            weights = None
+
+    if quantize:
+        weights = QuantizedResNeXt101_32x8dWeights.verify(weights)
+    else:
+        weights = ResNeXt101_32x8dWeights.verify(weights)
+
+    kwargs["groups"] = 32
+    kwargs["width_per_group"] = 8
+    return _resnet(QuantizableBottleneck, [3, 4, 23, 3], weights, progress, quantize, **kwargs)