diff --git a/test/test_extended_models.py b/test/test_extended_models.py index 1422cc28103..a07f741c9f7 100644 --- a/test/test_extended_models.py +++ b/test/test_extended_models.py @@ -81,7 +81,7 @@ def test_naming_conventions(model_fn): def test_schema_meta_validation(model_fn): classification_fields = ["size", "categories", "acc@1", "acc@5", "min_size"] defaults = { - "all": ["task", "architecture", "publication_year", "interpolation", "recipe", "num_params"], + "all": ["task", "architecture", "recipe", "num_params"], "models": classification_fields, "detection": ["categories", "map"], "quantization": classification_fields + ["backend", "quantization", "unquantized"], diff --git a/torchvision/models/alexnet.py b/torchvision/models/alexnet.py index 6ee5b98c673..0128b82b08a 100644 --- a/torchvision/models/alexnet.py +++ b/torchvision/models/alexnet.py @@ -4,7 +4,7 @@ import torch import torch.nn as nn -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -59,12 +59,10 @@ class AlexNet_Weights(WeightsEnum): meta={ "task": "image_classification", "architecture": "AlexNet", - "publication_year": 2012, "num_params": 61100840, "size": (224, 224), "min_size": (63, 63), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg", "acc@1": 56.522, "acc@5": 79.066, diff --git a/torchvision/models/convnext.py b/torchvision/models/convnext.py index 8774b9a1bc2..8720c590ede 100644 --- a/torchvision/models/convnext.py +++ b/torchvision/models/convnext.py @@ -7,7 +7,7 @@ from ..ops.misc import Conv2dNormActivation from ..ops.stochastic_depth import StochasticDepth -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -206,11 +206,9 @@ def _convnext( _COMMON_META = { "task": "image_classification", "architecture": "ConvNeXt", - "publication_year": 2022, "size": (224, 224), "min_size": (32, 32), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#convnext", } diff --git a/torchvision/models/densenet.py b/torchvision/models/densenet.py index 2ffb29c54cb..7bc53ad2679 100644 --- a/torchvision/models/densenet.py +++ b/torchvision/models/densenet.py @@ -9,7 +9,7 @@ import torch.utils.checkpoint as cp from torch import Tensor -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -268,11 +268,9 @@ def _densenet( _COMMON_META = { "task": "image_classification", "architecture": "DenseNet", - "publication_year": 2016, "size": (224, 224), "min_size": (29, 29), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/pull/116", } diff --git a/torchvision/models/detection/faster_rcnn.py b/torchvision/models/detection/faster_rcnn.py index 4334d3e4f67..ce794b0ed76 100644 --- a/torchvision/models/detection/faster_rcnn.py +++ b/torchvision/models/detection/faster_rcnn.py @@ -6,7 +6,7 @@ from torchvision.ops import MultiScaleRoIAlign from ...ops import misc as misc_nn_ops -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from .._api import WeightsEnum, Weights from .._meta import _COCO_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_value_param @@ -372,9 +372,7 @@ def forward(self, x): _COMMON_META = { "task": "image_object_detection", "architecture": "FasterRCNN", - "publication_year": 2015, "categories": _COCO_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } @@ -398,7 +396,6 @@ class FasterRCNN_ResNet50_FPN_V2_Weights(WeightsEnum): transforms=ObjectDetection, meta={ **_COMMON_META, - "publication_year": 2021, "num_params": 43712278, "recipe": "https://github.com/pytorch/vision/pull/5763", "map": 46.7, diff --git a/torchvision/models/detection/fcos.py b/torchvision/models/detection/fcos.py index 4e0e0424a0c..3bb9a35f517 100644 --- a/torchvision/models/detection/fcos.py +++ b/torchvision/models/detection/fcos.py @@ -11,7 +11,7 @@ from ...ops import boxes as box_ops from ...ops import misc as misc_nn_ops from ...ops.feature_pyramid_network import LastLevelP6P7 -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from ...utils import _log_api_usage_once from .._api import WeightsEnum, Weights from .._meta import _COCO_CATEGORIES @@ -653,10 +653,8 @@ class FCOS_ResNet50_FPN_Weights(WeightsEnum): meta={ "task": "image_object_detection", "architecture": "FCOS", - "publication_year": 2019, "num_params": 32269600, "categories": _COCO_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#fcos-resnet-50-fpn", "map": 39.2, }, diff --git a/torchvision/models/detection/keypoint_rcnn.py b/torchvision/models/detection/keypoint_rcnn.py index c7df4910009..469d97b3f2f 100644 --- a/torchvision/models/detection/keypoint_rcnn.py +++ b/torchvision/models/detection/keypoint_rcnn.py @@ -5,7 +5,7 @@ from torchvision.ops import MultiScaleRoIAlign from ...ops import misc as misc_nn_ops -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from .._api import WeightsEnum, Weights from .._meta import _COCO_PERSON_CATEGORIES, _COCO_PERSON_KEYPOINT_NAMES from .._utils import handle_legacy_interface, _ovewrite_value_param @@ -310,10 +310,8 @@ def forward(self, x): _COMMON_META = { "task": "image_object_detection", "architecture": "KeypointRCNN", - "publication_year": 2017, "categories": _COCO_PERSON_CATEGORIES, "keypoint_names": _COCO_PERSON_KEYPOINT_NAMES, - "interpolation": InterpolationMode.BILINEAR, } diff --git a/torchvision/models/detection/mask_rcnn.py b/torchvision/models/detection/mask_rcnn.py index 59ab8b0946f..ae6b0aa81f0 100644 --- a/torchvision/models/detection/mask_rcnn.py +++ b/torchvision/models/detection/mask_rcnn.py @@ -5,7 +5,7 @@ from torchvision.ops import MultiScaleRoIAlign from ...ops import misc as misc_nn_ops -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from .._api import WeightsEnum, Weights from .._meta import _COCO_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_value_param @@ -354,7 +354,6 @@ def __init__(self, in_channels, dim_reduced, num_classes): "task": "image_object_detection", "architecture": "MaskRCNN", "categories": _COCO_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } @@ -364,7 +363,6 @@ class MaskRCNN_ResNet50_FPN_Weights(WeightsEnum): transforms=ObjectDetection, meta={ **_COMMON_META, - "publication_year": 2017, "num_params": 44401393, "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn", "map": 37.9, @@ -380,7 +378,6 @@ class MaskRCNN_ResNet50_FPN_V2_Weights(WeightsEnum): transforms=ObjectDetection, meta={ **_COMMON_META, - "publication_year": 2021, "num_params": 46359409, "recipe": "https://github.com/pytorch/vision/pull/5773", "map": 47.4, diff --git a/torchvision/models/detection/retinanet.py b/torchvision/models/detection/retinanet.py index 954dfdba1a1..247e9cdc0d4 100644 --- a/torchvision/models/detection/retinanet.py +++ b/torchvision/models/detection/retinanet.py @@ -11,7 +11,7 @@ from ...ops import boxes as box_ops from ...ops import misc as misc_nn_ops from ...ops.feature_pyramid_network import LastLevelP6P7 -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from ...utils import _log_api_usage_once from .._api import WeightsEnum, Weights from .._meta import _COCO_CATEGORIES @@ -677,7 +677,6 @@ def forward(self, images, targets=None): "task": "image_object_detection", "architecture": "RetinaNet", "categories": _COCO_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } @@ -687,7 +686,6 @@ class RetinaNet_ResNet50_FPN_Weights(WeightsEnum): transforms=ObjectDetection, meta={ **_COMMON_META, - "publication_year": 2017, "num_params": 34014999, "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#retinanet", "map": 36.4, @@ -702,7 +700,6 @@ class RetinaNet_ResNet50_FPN_V2_Weights(WeightsEnum): transforms=ObjectDetection, meta={ **_COMMON_META, - "publication_year": 2019, "num_params": 38198935, "recipe": "https://github.com/pytorch/vision/pull/5756", "map": 41.5, diff --git a/torchvision/models/detection/ssd.py b/torchvision/models/detection/ssd.py index 537371fdc27..d4e26e1517b 100644 --- a/torchvision/models/detection/ssd.py +++ b/torchvision/models/detection/ssd.py @@ -7,7 +7,7 @@ from torch import nn, Tensor from ...ops import boxes as box_ops -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from ...utils import _log_api_usage_once from .._api import WeightsEnum, Weights from .._meta import _COCO_CATEGORIES @@ -32,11 +32,9 @@ class SSD300_VGG16_Weights(WeightsEnum): meta={ "task": "image_object_detection", "architecture": "SSD", - "publication_year": 2015, "num_params": 35641826, "size": (300, 300), "categories": _COCO_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#ssd300-vgg16", "map": 25.1, }, diff --git a/torchvision/models/detection/ssdlite.py b/torchvision/models/detection/ssdlite.py index 2e890356417..d55991f48d5 100644 --- a/torchvision/models/detection/ssdlite.py +++ b/torchvision/models/detection/ssdlite.py @@ -7,7 +7,7 @@ from torch import nn, Tensor from ...ops.misc import Conv2dNormActivation -from ...transforms._presets import ObjectDetection, InterpolationMode +from ...transforms._presets import ObjectDetection from ...utils import _log_api_usage_once from .. import mobilenet from .._api import WeightsEnum, Weights @@ -191,11 +191,9 @@ class SSDLite320_MobileNet_V3_Large_Weights(WeightsEnum): meta={ "task": "image_object_detection", "architecture": "SSDLite", - "publication_year": 2018, "num_params": 3440060, "size": (320, 320), "categories": _COCO_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#ssdlite320-mobilenetv3-large", "map": 21.3, }, diff --git a/torchvision/models/efficientnet.py b/torchvision/models/efficientnet.py index b9d3b9b30c9..a5519c60824 100644 --- a/torchvision/models/efficientnet.py +++ b/torchvision/models/efficientnet.py @@ -439,8 +439,6 @@ def _efficientnet_conf( _COMMON_META_V1 = { **_COMMON_META, "architecture": "EfficientNet", - "publication_year": 2019, - "interpolation": InterpolationMode.BICUBIC, "min_size": (1, 1), } @@ -448,8 +446,6 @@ def _efficientnet_conf( _COMMON_META_V2 = { **_COMMON_META, "architecture": "EfficientNetV2", - "publication_year": 2021, - "interpolation": InterpolationMode.BILINEAR, "min_size": (33, 33), } @@ -494,7 +490,6 @@ class EfficientNet_B1_Weights(WeightsEnum): **_COMMON_META_V1, "num_params": 7794184, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuning", - "interpolation": InterpolationMode.BILINEAR, "size": (240, 240), "acc@1": 79.838, "acc@5": 94.934, diff --git a/torchvision/models/googlenet.py b/torchvision/models/googlenet.py index ced92571974..a84d394d8d6 100644 --- a/torchvision/models/googlenet.py +++ b/torchvision/models/googlenet.py @@ -8,7 +8,7 @@ import torch.nn.functional as F from torch import Tensor -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -282,12 +282,10 @@ class GoogLeNet_Weights(WeightsEnum): meta={ "task": "image_classification", "architecture": "GoogLeNet", - "publication_year": 2014, "num_params": 6624904, "size": (224, 224), "min_size": (15, 15), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#googlenet", "acc@1": 69.778, "acc@5": 89.530, diff --git a/torchvision/models/inception.py b/torchvision/models/inception.py index 816fab45549..a1997eb8dbd 100644 --- a/torchvision/models/inception.py +++ b/torchvision/models/inception.py @@ -7,7 +7,7 @@ import torch.nn.functional as F from torch import nn, Tensor -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -414,12 +414,10 @@ class Inception_V3_Weights(WeightsEnum): meta={ "task": "image_classification", "architecture": "InceptionV3", - "publication_year": 2015, "num_params": 27161264, "size": (299, 299), "min_size": (75, 75), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#inception-v3", "acc@1": 77.294, "acc@5": 93.450, diff --git a/torchvision/models/mnasnet.py b/torchvision/models/mnasnet.py index 578e77f7934..7517d3c9772 100644 --- a/torchvision/models/mnasnet.py +++ b/torchvision/models/mnasnet.py @@ -6,7 +6,7 @@ import torch.nn as nn from torch import Tensor -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -214,11 +214,9 @@ def _load_from_state_dict( _COMMON_META = { "task": "image_classification", "architecture": "MNASNet", - "publication_year": 2018, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/1e100/mnasnet_trainer", } diff --git a/torchvision/models/mobilenetv2.py b/torchvision/models/mobilenetv2.py index 085049117ec..cf1580e3b75 100644 --- a/torchvision/models/mobilenetv2.py +++ b/torchvision/models/mobilenetv2.py @@ -7,7 +7,7 @@ from torch import nn from ..ops.misc import Conv2dNormActivation -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -197,12 +197,10 @@ def forward(self, x: Tensor) -> Tensor: _COMMON_META = { "task": "image_classification", "architecture": "MobileNetV2", - "publication_year": 2018, "num_params": 3504872, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } diff --git a/torchvision/models/mobilenetv3.py b/torchvision/models/mobilenetv3.py index 91e1ea91a94..8c0eb4c3223 100644 --- a/torchvision/models/mobilenetv3.py +++ b/torchvision/models/mobilenetv3.py @@ -6,7 +6,7 @@ from torch import nn, Tensor from ..ops.misc import Conv2dNormActivation, SqueezeExcitation as SElayer -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -306,11 +306,9 @@ def _mobilenet_v3( _COMMON_META = { "task": "image_classification", "architecture": "MobileNetV3", - "publication_year": 2019, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } diff --git a/torchvision/models/optical_flow/raft.py b/torchvision/models/optical_flow/raft.py index 244d2b2fac1..3a3d0334593 100644 --- a/torchvision/models/optical_flow/raft.py +++ b/torchvision/models/optical_flow/raft.py @@ -8,7 +8,7 @@ from torch.nn.modules.instancenorm import InstanceNorm2d from torchvision.ops import Conv2dNormActivation -from ...transforms._presets import OpticalFlow, InterpolationMode +from ...transforms._presets import OpticalFlow from ...utils import _log_api_usage_once from .._api import Weights, WeightsEnum from .._utils import handle_legacy_interface @@ -514,8 +514,6 @@ def forward(self, image1, image2, num_flow_updates: int = 12): _COMMON_META = { "task": "optical_flow", "architecture": "RAFT", - "publication_year": 2020, - "interpolation": InterpolationMode.BILINEAR, } diff --git a/torchvision/models/quantization/googlenet.py b/torchvision/models/quantization/googlenet.py index 1794c834eea..37a420a0a10 100644 --- a/torchvision/models/quantization/googlenet.py +++ b/torchvision/models/quantization/googlenet.py @@ -7,7 +7,7 @@ from torch import Tensor from torch.nn import functional as F -from ...transforms._presets import ImageClassification, InterpolationMode +from ...transforms._presets import ImageClassification from .._api import WeightsEnum, Weights from .._meta import _IMAGENET_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_named_param @@ -113,12 +113,10 @@ class GoogLeNet_QuantizedWeights(WeightsEnum): meta={ "task": "image_classification", "architecture": "GoogLeNet", - "publication_year": 2014, "num_params": 6624904, "size": (224, 224), "min_size": (15, 15), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "backend": "fbgemm", "quantization": "Post Training Quantization", "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models", diff --git a/torchvision/models/quantization/inception.py b/torchvision/models/quantization/inception.py index ff5c9a37365..288be0068ca 100644 --- a/torchvision/models/quantization/inception.py +++ b/torchvision/models/quantization/inception.py @@ -9,7 +9,7 @@ from torchvision.models import inception as inception_module from torchvision.models.inception import InceptionOutputs, Inception_V3_Weights -from ...transforms._presets import ImageClassification, InterpolationMode +from ...transforms._presets import ImageClassification from .._api import WeightsEnum, Weights from .._meta import _IMAGENET_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_named_param @@ -179,12 +179,10 @@ class Inception_V3_QuantizedWeights(WeightsEnum): meta={ "task": "image_classification", "architecture": "InceptionV3", - "publication_year": 2015, "num_params": 27161264, "size": (299, 299), "min_size": (75, 75), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "backend": "fbgemm", "quantization": "Post Training Quantization", "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models", diff --git a/torchvision/models/quantization/mobilenetv2.py b/torchvision/models/quantization/mobilenetv2.py index d9554e0ba9f..8c78ebcfd3b 100644 --- a/torchvision/models/quantization/mobilenetv2.py +++ b/torchvision/models/quantization/mobilenetv2.py @@ -7,7 +7,7 @@ from torchvision.models.mobilenetv2 import InvertedResidual, MobileNetV2, MobileNet_V2_Weights from ...ops.misc import Conv2dNormActivation -from ...transforms._presets import ImageClassification, InterpolationMode +from ...transforms._presets import ImageClassification from .._api import WeightsEnum, Weights from .._meta import _IMAGENET_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_named_param @@ -71,12 +71,10 @@ class MobileNet_V2_QuantizedWeights(WeightsEnum): meta={ "task": "image_classification", "architecture": "MobileNetV2", - "publication_year": 2018, "num_params": 3504872, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "backend": "qnnpack", "quantization": "Quantization Aware Training", "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv2", diff --git a/torchvision/models/quantization/mobilenetv3.py b/torchvision/models/quantization/mobilenetv3.py index 88907ec210a..afee0b1d954 100644 --- a/torchvision/models/quantization/mobilenetv3.py +++ b/torchvision/models/quantization/mobilenetv3.py @@ -6,7 +6,7 @@ from torch.ao.quantization import QuantStub, DeQuantStub from ...ops.misc import Conv2dNormActivation, SqueezeExcitation -from ...transforms._presets import ImageClassification, InterpolationMode +from ...transforms._presets import ImageClassification from .._api import WeightsEnum, Weights from .._meta import _IMAGENET_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_named_param @@ -161,12 +161,10 @@ class MobileNet_V3_Large_QuantizedWeights(WeightsEnum): meta={ "task": "image_classification", "architecture": "MobileNetV3", - "publication_year": 2019, "num_params": 5483032, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "backend": "qnnpack", "quantization": "Quantization Aware Training", "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv3", diff --git a/torchvision/models/quantization/resnet.py b/torchvision/models/quantization/resnet.py index a781f320000..4534678af7f 100644 --- a/torchvision/models/quantization/resnet.py +++ b/torchvision/models/quantization/resnet.py @@ -13,7 +13,7 @@ ResNeXt101_32X8D_Weights, ) -from ...transforms._presets import ImageClassification, InterpolationMode +from ...transforms._presets import ImageClassification from .._api import WeightsEnum, Weights from .._meta import _IMAGENET_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_named_param @@ -151,7 +151,6 @@ def _resnet( "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "backend": "fbgemm", "quantization": "Post Training Quantization", "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models", @@ -165,7 +164,6 @@ class ResNet18_QuantizedWeights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 11689512, "unquantized": ResNet18_Weights.IMAGENET1K_V1, "acc@1": 69.494, @@ -182,7 +180,6 @@ class ResNet50_QuantizedWeights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 25557032, "unquantized": ResNet50_Weights.IMAGENET1K_V1, "acc@1": 75.920, @@ -195,7 +192,6 @@ class ResNet50_QuantizedWeights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 25557032, "unquantized": ResNet50_Weights.IMAGENET1K_V2, "acc@1": 80.282, @@ -212,7 +208,6 @@ class ResNeXt101_32X8D_QuantizedWeights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNeXt", - "publication_year": 2016, "num_params": 88791336, "unquantized": ResNeXt101_32X8D_Weights.IMAGENET1K_V1, "acc@1": 78.986, @@ -225,7 +220,6 @@ class ResNeXt101_32X8D_QuantizedWeights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNeXt", - "publication_year": 2016, "num_params": 88791336, "unquantized": ResNeXt101_32X8D_Weights.IMAGENET1K_V2, "acc@1": 82.574, diff --git a/torchvision/models/quantization/shufflenetv2.py b/torchvision/models/quantization/shufflenetv2.py index 1f4f1890e07..5e679da7399 100644 --- a/torchvision/models/quantization/shufflenetv2.py +++ b/torchvision/models/quantization/shufflenetv2.py @@ -6,7 +6,7 @@ from torch import Tensor from torchvision.models import shufflenetv2 -from ...transforms._presets import ImageClassification, InterpolationMode +from ...transforms._presets import ImageClassification from .._api import WeightsEnum, Weights from .._meta import _IMAGENET_CATEGORIES from .._utils import handle_legacy_interface, _ovewrite_named_param @@ -104,11 +104,9 @@ def _shufflenetv2( _COMMON_META = { "task": "image_classification", "architecture": "ShuffleNetV2", - "publication_year": 2018, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "backend": "fbgemm", "quantization": "Post Training Quantization", "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models", diff --git a/torchvision/models/regnet.py b/torchvision/models/regnet.py index 9775a92a300..17c54203a8f 100644 --- a/torchvision/models/regnet.py +++ b/torchvision/models/regnet.py @@ -405,20 +405,16 @@ def _regnet( _COMMON_META = { "task": "image_classification", "architecture": "RegNet", - "publication_year": 2020, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } _COMMON_SWAG_META = { **_COMMON_META, - "publication_year": 2022, "size": (384, 384), "recipe": "https://github.com/facebookresearch/SWAG", "license": "https://github.com/facebookresearch/SWAG/blob/main/LICENSE", - "interpolation": InterpolationMode.BICUBIC, } diff --git a/torchvision/models/resnet.py b/torchvision/models/resnet.py index 3d1a831becf..c0c102710be 100644 --- a/torchvision/models/resnet.py +++ b/torchvision/models/resnet.py @@ -5,7 +5,7 @@ import torch.nn as nn from torch import Tensor -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -306,7 +306,6 @@ def _resnet( "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } @@ -317,7 +316,6 @@ class ResNet18_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 11689512, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnet", "acc@1": 69.758, @@ -334,7 +332,6 @@ class ResNet34_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 21797672, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnet", "acc@1": 73.314, @@ -351,7 +348,6 @@ class ResNet50_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 25557032, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnet", "acc@1": 76.130, @@ -364,7 +360,6 @@ class ResNet50_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 25557032, "recipe": "https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621", "acc@1": 80.858, @@ -381,7 +376,6 @@ class ResNet101_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 44549160, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnet", "acc@1": 77.374, @@ -394,7 +388,6 @@ class ResNet101_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 44549160, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe", "acc@1": 81.886, @@ -411,7 +404,6 @@ class ResNet152_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 60192808, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnet", "acc@1": 78.312, @@ -424,7 +416,6 @@ class ResNet152_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNet", - "publication_year": 2015, "num_params": 60192808, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe", "acc@1": 82.284, @@ -441,7 +432,6 @@ class ResNeXt50_32X4D_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNeXt", - "publication_year": 2016, "num_params": 25028904, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnext", "acc@1": 77.618, @@ -454,7 +444,6 @@ class ResNeXt50_32X4D_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNeXt", - "publication_year": 2016, "num_params": 25028904, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe", "acc@1": 81.198, @@ -471,7 +460,6 @@ class ResNeXt101_32X8D_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNeXt", - "publication_year": 2016, "num_params": 88791336, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#resnext", "acc@1": 79.312, @@ -484,7 +472,6 @@ class ResNeXt101_32X8D_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "ResNeXt", - "publication_year": 2016, "num_params": 88791336, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres", "acc@1": 82.834, @@ -501,7 +488,6 @@ class Wide_ResNet50_2_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "WideResNet", - "publication_year": 2016, "num_params": 68883240, "recipe": "https://github.com/pytorch/vision/pull/912#issue-445437439", "acc@1": 78.468, @@ -514,7 +500,6 @@ class Wide_ResNet50_2_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "WideResNet", - "publication_year": 2016, "num_params": 68883240, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres", "acc@1": 81.602, @@ -531,7 +516,6 @@ class Wide_ResNet101_2_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "WideResNet", - "publication_year": 2016, "num_params": 126886696, "recipe": "https://github.com/pytorch/vision/pull/912#issue-445437439", "acc@1": 78.848, @@ -544,7 +528,6 @@ class Wide_ResNet101_2_Weights(WeightsEnum): meta={ **_COMMON_META, "architecture": "WideResNet", - "publication_year": 2016, "num_params": 126886696, "recipe": "https://github.com/pytorch/vision/issues/3995#new-recipe", "acc@1": 82.510, diff --git a/torchvision/models/segmentation/deeplabv3.py b/torchvision/models/segmentation/deeplabv3.py index 092a81f643b..beaa1a8c7fe 100644 --- a/torchvision/models/segmentation/deeplabv3.py +++ b/torchvision/models/segmentation/deeplabv3.py @@ -5,7 +5,7 @@ from torch import nn from torch.nn import functional as F -from ...transforms._presets import SemanticSegmentation, InterpolationMode +from ...transforms._presets import SemanticSegmentation from .._api import WeightsEnum, Weights from .._meta import _VOC_CATEGORIES from .._utils import IntermediateLayerGetter, handle_legacy_interface, _ovewrite_value_param @@ -131,9 +131,7 @@ def _deeplabv3_resnet( _COMMON_META = { "task": "image_semantic_segmentation", "architecture": "DeepLabV3", - "publication_year": 2017, "categories": _VOC_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } diff --git a/torchvision/models/segmentation/fcn.py b/torchvision/models/segmentation/fcn.py index 6b6d14ffe32..678872915eb 100644 --- a/torchvision/models/segmentation/fcn.py +++ b/torchvision/models/segmentation/fcn.py @@ -3,7 +3,7 @@ from torch import nn -from ...transforms._presets import SemanticSegmentation, InterpolationMode +from ...transforms._presets import SemanticSegmentation from .._api import WeightsEnum, Weights from .._meta import _VOC_CATEGORIES from .._utils import IntermediateLayerGetter, handle_legacy_interface, _ovewrite_value_param @@ -50,9 +50,7 @@ def __init__(self, in_channels: int, channels: int) -> None: _COMMON_META = { "task": "image_semantic_segmentation", "architecture": "FCN", - "publication_year": 2014, "categories": _VOC_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } diff --git a/torchvision/models/segmentation/lraspp.py b/torchvision/models/segmentation/lraspp.py index fc6d14d366b..eb66fe93af2 100644 --- a/torchvision/models/segmentation/lraspp.py +++ b/torchvision/models/segmentation/lraspp.py @@ -5,7 +5,7 @@ from torch import nn, Tensor from torch.nn import functional as F -from ...transforms._presets import SemanticSegmentation, InterpolationMode +from ...transforms._presets import SemanticSegmentation from ...utils import _log_api_usage_once from .._api import WeightsEnum, Weights from .._meta import _VOC_CATEGORIES @@ -100,10 +100,8 @@ class LRASPP_MobileNet_V3_Large_Weights(WeightsEnum): meta={ "task": "image_semantic_segmentation", "architecture": "LRASPP", - "publication_year": 2019, "num_params": 3221538, "categories": _VOC_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_large", "mIoU": 57.9, "acc": 91.2, diff --git a/torchvision/models/shufflenetv2.py b/torchvision/models/shufflenetv2.py index e988b819078..4bde891d6b0 100644 --- a/torchvision/models/shufflenetv2.py +++ b/torchvision/models/shufflenetv2.py @@ -5,7 +5,7 @@ import torch.nn as nn from torch import Tensor -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -186,11 +186,9 @@ def _shufflenetv2( _COMMON_META = { "task": "image_classification", "architecture": "ShuffleNetV2", - "publication_year": 2018, "size": (224, 224), "min_size": (1, 1), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/barrh/Shufflenet-v2-Pytorch/tree/v0.1.0", } diff --git a/torchvision/models/squeezenet.py b/torchvision/models/squeezenet.py index 419fc892131..42618f02b9d 100644 --- a/torchvision/models/squeezenet.py +++ b/torchvision/models/squeezenet.py @@ -5,7 +5,7 @@ import torch.nn as nn import torch.nn.init as init -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -117,10 +117,8 @@ def _squeezenet( _COMMON_META = { "task": "image_classification", "architecture": "SqueezeNet", - "publication_year": 2016, "size": (224, 224), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/pull/49#issuecomment-277560717", } diff --git a/torchvision/models/vgg.py b/torchvision/models/vgg.py index 45f2dae5808..e75c8ad2c6f 100644 --- a/torchvision/models/vgg.py +++ b/torchvision/models/vgg.py @@ -4,7 +4,7 @@ import torch import torch.nn as nn -from ..transforms._presets import ImageClassification, InterpolationMode +from ..transforms._presets import ImageClassification from ..utils import _log_api_usage_once from ._api import WeightsEnum, Weights from ._meta import _IMAGENET_CATEGORIES @@ -109,11 +109,9 @@ def _vgg(cfg: str, batch_norm: bool, weights: Optional[WeightsEnum], progress: b _COMMON_META = { "task": "image_classification", "architecture": "VGG", - "publication_year": 2014, "size": (224, 224), "min_size": (32, 32), "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg", } diff --git a/torchvision/models/video/resnet.py b/torchvision/models/video/resnet.py index 618ddb96ba2..b0fcd39fecc 100644 --- a/torchvision/models/video/resnet.py +++ b/torchvision/models/video/resnet.py @@ -4,7 +4,7 @@ import torch.nn as nn from torch import Tensor -from ...transforms._presets import VideoClassification, InterpolationMode +from ...transforms._presets import VideoClassification from ...utils import _log_api_usage_once from .._api import WeightsEnum, Weights from .._meta import _KINETICS400_CATEGORIES @@ -310,11 +310,9 @@ def _video_resnet( _COMMON_META = { "task": "video_classification", - "publication_year": 2017, "size": (112, 112), "min_size": (1, 1), "categories": _KINETICS400_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, "recipe": "https://github.com/pytorch/vision/tree/main/references/video_classification", } diff --git a/torchvision/models/vision_transformer.py b/torchvision/models/vision_transformer.py index 9837f9b81b4..ffc6265eafc 100644 --- a/torchvision/models/vision_transformer.py +++ b/torchvision/models/vision_transformer.py @@ -1,7 +1,7 @@ import math from collections import OrderedDict from functools import partial -from typing import Any, Callable, List, NamedTuple, Optional, Sequence +from typing import Any, Callable, List, NamedTuple, Optional, Sequence, Dict import torch import torch.nn as nn @@ -318,20 +318,16 @@ def _vision_transformer( return model -_COMMON_META = { +_COMMON_META: Dict[str, Any] = { "task": "image_classification", "architecture": "ViT", - "publication_year": 2020, "categories": _IMAGENET_CATEGORIES, - "interpolation": InterpolationMode.BILINEAR, } -_COMMON_SWAG_META = { +_COMMON_SWAG_META: Dict[str, Any] = { **_COMMON_META, - "publication_year": 2022, "recipe": "https://github.com/facebookresearch/SWAG", "license": "https://github.com/facebookresearch/SWAG/blob/main/LICENSE", - "interpolation": InterpolationMode.BICUBIC, }