remove functionality scheduled for 0.15 after deprecation (#7176)

torchvision/models/quantization/mobilenetv2.py

@@ -152,15 +152,3 @@ def mobilenet_v2(
         model.load_state_dict(weights.get_state_dict(progress=progress))
 
     return model
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-from ..mobilenetv2 import model_urls  # noqa: F401
-
-
-quant_model_urls = _ModelURLs(
-    {
-        "mobilenet_v2_qnnpack": MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1.url,
-    }
-)

torchvision/models/quantization/mobilenetv3.py

@@ -235,15 +235,3 @@ def mobilenet_v3_large(
 
     inverted_residual_setting, last_channel = _mobilenet_v3_conf("mobilenet_v3_large", **kwargs)
     return _mobilenet_v3_model(inverted_residual_setting, last_channel, weights, progress, quantize, **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-from ..mobilenetv3 import model_urls  # noqa: F401
-
-
-quant_model_urls = _ModelURLs(
-    {
-        "mobilenet_v3_large_qnnpack": MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1.url,
-    }
-)

torchvision/models/quantization/resnet.py

@@ -482,17 +482,3 @@ def resnext101_64x4d(
     _ovewrite_named_param(kwargs, "groups", 64)
     _ovewrite_named_param(kwargs, "width_per_group", 4)
     return _resnet(QuantizableBottleneck, [3, 4, 23, 3], weights, progress, quantize, **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-from ..resnet import model_urls  # noqa: F401
-
-
-quant_model_urls = _ModelURLs(
-    {
-        "resnet18_fbgemm": ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1.url,
-        "resnet50_fbgemm": ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V1.url,
-        "resnext101_32x8d_fbgemm": ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V1.url,
-    }
-)

torchvision/models/quantization/shufflenetv2.py

@@ -425,16 +425,3 @@ def shufflenet_v2_x2_0(
     return _shufflenetv2(
         [4, 8, 4], [24, 244, 488, 976, 2048], weights=weights, progress=progress, quantize=quantize, **kwargs
     )
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-from ..shufflenetv2 import model_urls  # noqa: F401
-
-
-quant_model_urls = _ModelURLs(
-    {
-        "shufflenetv2_x0.5_fbgemm": ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1.url,
-        "shufflenetv2_x1.0_fbgemm": ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1.url,
-    }
-)

torchvision/models/regnet.py

@@ -1569,27 +1569,3 @@ def regnet_x_32gf(*, weights: Optional[RegNet_X_32GF_Weights] = None, progress:
 
     params = BlockParams.from_init_params(depth=23, w_0=320, w_a=69.86, w_m=2.0, group_width=168, **kwargs)
     return _regnet(params, weights, progress, **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from ._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "regnet_y_400mf": RegNet_Y_400MF_Weights.IMAGENET1K_V1.url,
-        "regnet_y_800mf": RegNet_Y_800MF_Weights.IMAGENET1K_V1.url,
-        "regnet_y_1_6gf": RegNet_Y_1_6GF_Weights.IMAGENET1K_V1.url,
-        "regnet_y_3_2gf": RegNet_Y_3_2GF_Weights.IMAGENET1K_V1.url,
-        "regnet_y_8gf": RegNet_Y_8GF_Weights.IMAGENET1K_V1.url,
-        "regnet_y_16gf": RegNet_Y_16GF_Weights.IMAGENET1K_V1.url,
-        "regnet_y_32gf": RegNet_Y_32GF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_400mf": RegNet_X_400MF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_800mf": RegNet_X_800MF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_1_6gf": RegNet_X_1_6GF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_3_2gf": RegNet_X_3_2GF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_8gf": RegNet_X_8GF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_16gf": RegNet_X_16GF_Weights.IMAGENET1K_V1.url,
-        "regnet_x_32gf": RegNet_X_32GF_Weights.IMAGENET1K_V1.url,
-    }
-)

torchvision/models/resnet.py

@@ -983,22 +983,3 @@ def wide_resnet101_2(
 
     _ovewrite_named_param(kwargs, "width_per_group", 64 * 2)
     return _resnet(Bottleneck, [3, 4, 23, 3], weights, progress, **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from ._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "resnet18": ResNet18_Weights.IMAGENET1K_V1.url,
-        "resnet34": ResNet34_Weights.IMAGENET1K_V1.url,
-        "resnet50": ResNet50_Weights.IMAGENET1K_V1.url,
-        "resnet101": ResNet101_Weights.IMAGENET1K_V1.url,
-        "resnet152": ResNet152_Weights.IMAGENET1K_V1.url,
-        "resnext50_32x4d": ResNeXt50_32X4D_Weights.IMAGENET1K_V1.url,
-        "resnext101_32x8d": ResNeXt101_32X8D_Weights.IMAGENET1K_V1.url,
-        "wide_resnet50_2": Wide_ResNet50_2_Weights.IMAGENET1K_V1.url,
-        "wide_resnet101_2": Wide_ResNet101_2_Weights.IMAGENET1K_V1.url,
-    }
-)

torchvision/models/segmentation/deeplabv3.py

@@ -388,16 +388,3 @@ def deeplabv3_mobilenet_v3_large(
         model.load_state_dict(weights.get_state_dict(progress=progress))
 
     return model
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "deeplabv3_resnet50_coco": DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1.url,
-        "deeplabv3_resnet101_coco": DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1.url,
-        "deeplabv3_mobilenet_v3_large_coco": DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1.url,
-    }
-)

torchvision/models/segmentation/fcn.py

@@ -230,15 +230,3 @@ def fcn_resnet101(
         model.load_state_dict(weights.get_state_dict(progress=progress))
 
     return model
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "fcn_resnet50_coco": FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1.url,
-        "fcn_resnet101_coco": FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1.url,
-    }
-)

torchvision/models/segmentation/lraspp.py

@@ -176,14 +176,3 @@ def lraspp_mobilenet_v3_large(
         model.load_state_dict(weights.get_state_dict(progress=progress))
 
     return model
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from .._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "lraspp_mobilenet_v3_large_coco": LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1.url,
-    }
-)

torchvision/models/shufflenetv2.py

@@ -406,17 +406,3 @@ def shufflenet_v2_x2_0(
     weights = ShuffleNet_V2_X2_0_Weights.verify(weights)
 
     return _shufflenetv2(weights, progress, [4, 8, 4], [24, 244, 488, 976, 2048], **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from ._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "shufflenetv2_x0.5": ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1.url,
-        "shufflenetv2_x1.0": ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1.url,
-        "shufflenetv2_x1.5": None,
-        "shufflenetv2_x2.0": None,
-    }
-)

torchvision/models/squeezenet.py

@@ -221,15 +221,3 @@ def squeezenet1_1(
     """
     weights = SqueezeNet1_1_Weights.verify(weights)
     return _squeezenet("1_1", weights, progress, **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from ._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "squeezenet1_0": SqueezeNet1_0_Weights.IMAGENET1K_V1.url,
-        "squeezenet1_1": SqueezeNet1_1_Weights.IMAGENET1K_V1.url,
-    }
-)

torchvision/models/vgg.py

@@ -509,21 +509,3 @@ def vgg19_bn(*, weights: Optional[VGG19_BN_Weights] = None, progress: bool = Tru
     weights = VGG19_BN_Weights.verify(weights)
 
     return _vgg("E", True, weights, progress, **kwargs)
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from ._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "vgg11": VGG11_Weights.IMAGENET1K_V1.url,
-        "vgg13": VGG13_Weights.IMAGENET1K_V1.url,
-        "vgg16": VGG16_Weights.IMAGENET1K_V1.url,
-        "vgg19": VGG19_Weights.IMAGENET1K_V1.url,
-        "vgg11_bn": VGG11_BN_Weights.IMAGENET1K_V1.url,
-        "vgg13_bn": VGG13_BN_Weights.IMAGENET1K_V1.url,
-        "vgg16_bn": VGG16_BN_Weights.IMAGENET1K_V1.url,
-        "vgg19_bn": VGG19_BN_Weights.IMAGENET1K_V1.url,
-    }
-)

torchvision/models/vision_transformer.py

@@ -862,17 +862,3 @@ def interpolate_embeddings(
             model_state = model_state_copy
 
     return model_state
-
-
-# The dictionary below is internal implementation detail and will be removed in v0.15
-from ._utils import _ModelURLs
-
-
-model_urls = _ModelURLs(
-    {
-        "vit_b_16": ViT_B_16_Weights.IMAGENET1K_V1.url,
-        "vit_b_32": ViT_B_32_Weights.IMAGENET1K_V1.url,
-        "vit_l_16": ViT_L_16_Weights.IMAGENET1K_V1.url,
-        "vit_l_32": ViT_L_32_Weights.IMAGENET1K_V1.url,
-    }
-)

torchvision/transforms/functional.py

@@ -421,8 +421,6 @@ def resize(
             Default is ``InterpolationMode.BILINEAR``. If input is Tensor, only ``InterpolationMode.NEAREST``,
             ``InterpolationMode.NEAREST_EXACT``, ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are
             supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         max_size (int, optional): The maximum allowed for the longer edge of
             the resized image: if the longer edge of the image is greater
             than ``max_size`` after being resized according to ``size``, then
@@ -441,13 +439,6 @@ def resize(
     """
     if not torch.jit.is_scripting() and not torch.jit.is_tracing():
         _log_api_usage_once(resize)
-    # Backward compatibility with integer value
-    if isinstance(interpolation, int):
-        warnings.warn(
-            "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-            "Please use InterpolationMode enum."
-        )
-        interpolation = _interpolation_modes_from_int(interpolation)
 
     if not isinstance(interpolation, InterpolationMode):
         raise TypeError("Argument interpolation should be a InterpolationMode")
@@ -623,8 +614,6 @@ def resized_crop(
             Default is ``InterpolationMode.BILINEAR``. If input is Tensor, only ``InterpolationMode.NEAREST``,
             ``InterpolationMode.NEAREST_EXACT``, ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are
             supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         antialias (bool, optional): antialias flag. If ``img`` is PIL Image, the flag is ignored and anti-alias
             is always used. If ``img`` is Tensor, the flag is False by default and can be set to True for
             ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` modes.
@@ -707,8 +696,6 @@ def perspective(
         interpolation (InterpolationMode): Desired interpolation enum defined by
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         fill (sequence or number, optional): Pixel fill value for the area outside the transformed
             image. If given a number, the value is used for all bands respectively.
 
@@ -724,14 +711,6 @@ def perspective(
 
     coeffs = _get_perspective_coeffs(startpoints, endpoints)
 
-    # Backward compatibility with integer value
-    if isinstance(interpolation, int):
-        warnings.warn(
-            "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-            "Please use InterpolationMode enum."
-        )
-        interpolation = _interpolation_modes_from_int(interpolation)
-
     if not isinstance(interpolation, InterpolationMode):
         raise TypeError("Argument interpolation should be a InterpolationMode")
 
@@ -1067,8 +1046,6 @@ def rotate(
         interpolation (InterpolationMode): Desired interpolation enum defined by
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         expand (bool, optional): Optional expansion flag.
             If true, expands the output image to make it large enough to hold the entire rotated image.
             If false or omitted, make the output image the same size as the input image.
@@ -1090,14 +1067,6 @@ def rotate(
     if not torch.jit.is_scripting() and not torch.jit.is_tracing():
         _log_api_usage_once(rotate)
 
-    # Backward compatibility with integer value
-    if isinstance(interpolation, int):
-        warnings.warn(
-            "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-            "Please use InterpolationMode enum."
-        )
-        interpolation = _interpolation_modes_from_int(interpolation)
-
     if not isinstance(angle, (int, float)):
         raise TypeError("Argument angle should be int or float")
 
@@ -1148,8 +1117,6 @@ def affine(
         interpolation (InterpolationMode): Desired interpolation enum defined by
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         fill (sequence or number, optional): Pixel fill value for the area outside the transformed
             image. If given a number, the value is used for all bands respectively.
 
@@ -1165,14 +1132,6 @@ def affine(
     if not torch.jit.is_scripting() and not torch.jit.is_tracing():
         _log_api_usage_once(affine)
 
-    # Backward compatibility with integer value
-    if isinstance(interpolation, int):
-        warnings.warn(
-            "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-            "Please use InterpolationMode enum."
-        )
-        interpolation = _interpolation_modes_from_int(interpolation)
-
     if not isinstance(angle, (int, float)):
         raise TypeError("Argument angle should be int or float")
 

torchvision/transforms/transforms.py

@@ -298,8 +298,6 @@ class Resize(torch.nn.Module):
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,
             ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         max_size (int, optional): The maximum allowed for the longer edge of
             the resized image: if the longer edge of the image is greater
             than ``max_size`` after being resized according to ``size``, then
@@ -324,14 +322,6 @@ class Resize(torch.nn.Module):
         self.size = size
         self.max_size = max_size
 
-        # Backward compatibility with integer value
-        if isinstance(interpolation, int):
-            warnings.warn(
-                "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-                "Please use InterpolationMode enum."
-            )
-            interpolation = _interpolation_modes_from_int(interpolation)
-
         self.interpolation = interpolation
         self.antialias = antialias
 
@@ -752,8 +742,6 @@ class RandomPerspective(torch.nn.Module):
         interpolation (InterpolationMode): Desired interpolation enum defined by
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         fill (sequence or number): Pixel fill value for the area outside the transformed
             image. Default is ``0``. If given a number, the value is used for all bands respectively.
     """
@@ -763,14 +751,6 @@ class RandomPerspective(torch.nn.Module):
         _log_api_usage_once(self)
         self.p = p
 
-        # Backward compatibility with integer value
-        if isinstance(interpolation, int):
-            warnings.warn(
-                "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-                "Please use InterpolationMode enum."
-            )
-            interpolation = _interpolation_modes_from_int(interpolation)
-
         self.interpolation = interpolation
         self.distortion_scale = distortion_scale
 
@@ -867,8 +847,6 @@ class RandomResizedCrop(torch.nn.Module):
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,
             ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         antialias (bool, optional): antialias flag. If ``img`` is PIL Image, the flag is ignored and anti-alias
             is always used. If ``img`` is Tensor, the flag is False by default and can be set to True for
             ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` modes.
@@ -894,14 +872,6 @@ class RandomResizedCrop(torch.nn.Module):
         if (scale[0] > scale[1]) or (ratio[0] > ratio[1]):
             warnings.warn("Scale and ratio should be of kind (min, max)")
 
-        # Backward compatibility with integer value
-        if isinstance(interpolation, int):
-            warnings.warn(
-                "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-                "Please use InterpolationMode enum."
-            )
-            interpolation = _interpolation_modes_from_int(interpolation)
-
         self.interpolation = interpolation
         self.antialias = antialias
         self.scale = scale
@@ -1288,8 +1258,6 @@ class RandomRotation(torch.nn.Module):
         interpolation (InterpolationMode): Desired interpolation enum defined by
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         expand (bool, optional): Optional expansion flag.
             If true, expands the output to make it large enough to hold the entire rotated image.
             If false or omitted, make the output image the same size as the input image.
@@ -1307,14 +1275,6 @@ class RandomRotation(torch.nn.Module):
         super().__init__()
         _log_api_usage_once(self)
 
-        # Backward compatibility with integer value
-        if isinstance(interpolation, int):
-            warnings.warn(
-                "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-                "Please use InterpolationMode enum."
-            )
-            interpolation = _interpolation_modes_from_int(interpolation)
-
         self.degrees = _setup_angle(degrees, name="degrees", req_sizes=(2,))
 
         if center is not None:
@@ -1398,8 +1358,6 @@ class RandomAffine(torch.nn.Module):
         interpolation (InterpolationMode): Desired interpolation enum defined by
             :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.
             If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.
-            For backward compatibility integer values (e.g. ``PIL.Image[.Resampling].NEAREST``) are still accepted,
-            but deprecated since 0.13 and will be removed in 0.15. Please use InterpolationMode enum.
         fill (sequence or number): Pixel fill value for the area outside the transformed
             image. Default is ``0``. If given a number, the value is used for all bands respectively.
         center (sequence, optional): Optional center of rotation, (x, y). Origin is the upper left corner.
@@ -1422,14 +1380,6 @@ class RandomAffine(torch.nn.Module):
         super().__init__()
         _log_api_usage_once(self)
 
-        # Backward compatibility with integer value
-        if isinstance(interpolation, int):
-            warnings.warn(
-                "Argument 'interpolation' of type int is deprecated since 0.13 and will be removed in 0.15. "
-                "Please use InterpolationMode enum."
-            )
-            interpolation = _interpolation_modes_from_int(interpolation)
-
         self.degrees = _setup_angle(degrees, name="degrees", req_sizes=(2,))
 
         if translate is not None: