Document all pre-trained Classification weights

docs/source/conf.py

@@ -335,8 +335,6 @@ def inject_weight_metadata(app, what, name, obj, options, lines):
 
         for field in obj:
             lines += [f"**{str(field)}**:", ""]
-            if field == obj.DEFAULT:
-                lines += [f"This weight is also available as ``{obj.__name__}.DEFAULT``.", ""]
 
             table = []
 
@@ -349,7 +347,12 @@ def inject_weight_metadata(app, what, name, obj, options, lines):
 
             custom_docs = meta_with_metrics.pop("_docs", None)  # Custom per-Weights docs
             if custom_docs is not None:
-                lines += [custom_docs, ""]
+                lines += [custom_docs]
+
+            if field == obj.DEFAULT:
+                lines += [f"Also available as ``{obj.__name__}.DEFAULT``."]
+
+            lines += [""]
 
             for k, v in meta_with_metrics.items():
                 if k in {"recipe", "license"}:
@@ -367,8 +370,8 @@ def inject_weight_metadata(app, what, name, obj, options, lines):
             lines += textwrap.indent(table, " " * 4).split("\n")
             lines.append("")
             lines.append(
-                f"The preprocessing/inference transforms are available at ``{str(field)}.transforms`` and "
-                f"perform the following operations: {field.transforms().describe()}"
+                f"The inference transforms are available at ``{str(field)}.transforms`` and "
+                f"perform the following preprocessing operations: {field.transforms().describe()}"
             )
             lines.append("")
 

test/test_extended_models.py

@@ -96,7 +96,7 @@ def test_schema_meta_validation(model_fn):
     classification_fields = {"categories", ("metrics", "acc@1"), ("metrics", "acc@5")}
     defaults = {
         "all": {"metrics", "min_size", "num_params", "recipe"},
-        "models": classification_fields,
+        "models": classification_fields | {"_docs"},
         "detection": {"categories", ("metrics", "box_map")},
         "quantization": classification_fields | {"backend", "unquantized"},
         "segmentation": {"categories", ("metrics", "miou"), ("metrics", "pixel_acc")},

torchvision/models/alexnet.py

@@ -65,6 +65,9 @@ class AlexNet_Weights(WeightsEnum):
                 "acc@1": 56.522,
                 "acc@5": 79.066,
             },
+            "_docs": """
+                These weights reproduce closely the results of the paper using a simplified training recipe.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V1

torchvision/models/convnext.py

@@ -207,6 +207,11 @@ def _convnext(
     "min_size": (32, 32),
     "categories": _IMAGENET_CATEGORIES,
     "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#convnext",
+    "_docs": """
+        These weights improve upon the results of the original paper by using a modified version of TorchVision's
+        `new training recipe
+        <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+    """,
 }
 
 

torchvision/models/densenet.py

@@ -261,6 +261,7 @@ def _densenet(
     "min_size": (29, 29),
     "categories": _IMAGENET_CATEGORIES,
     "recipe": "https://github.com/pytorch/vision/pull/116",
+    "_docs": """These weights are ported from LuaTorch.""",
 }
 
 

torchvision/models/efficientnet.py

@@ -431,24 +431,26 @@ def _efficientnet_conf(
 
 _COMMON_META: Dict[str, Any] = {
     "categories": _IMAGENET_CATEGORIES,
-    "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#efficientnet",
 }
 
 
 _COMMON_META_V1 = {
     **_COMMON_META,
     "min_size": (1, 1),
+    "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1",
 }
 
 
 _COMMON_META_V2 = {
     **_COMMON_META,
     "min_size": (33, 33),
+    "recipe": "https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2",
 }
 
 
 class EfficientNet_B0_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/rwightman/pytorch-image-models/
         url="https://download.pytorch.org/models/efficientnet_b0_rwightman-3dd342df.pth",
         transforms=partial(
             ImageClassification, crop_size=224, resize_size=256, interpolation=InterpolationMode.BICUBIC
@@ -460,13 +462,15 @@ class EfficientNet_B0_Weights(WeightsEnum):
                 "acc@1": 77.692,
                 "acc@5": 93.532,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_B1_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/rwightman/pytorch-image-models/
         url="https://download.pytorch.org/models/efficientnet_b1_rwightman-533bc792.pth",
         transforms=partial(
             ImageClassification, crop_size=240, resize_size=256, interpolation=InterpolationMode.BICUBIC
@@ -478,6 +482,7 @@ class EfficientNet_B1_Weights(WeightsEnum):
                 "acc@1": 78.642,
                 "acc@5": 94.186,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     IMAGENET1K_V2 = Weights(
@@ -493,13 +498,19 @@ class EfficientNet_B1_Weights(WeightsEnum):
                 "acc@1": 79.838,
                 "acc@5": 94.934,
             },
+            "_docs": """
+                These weights improve upon the results of the original paper by using a modified version of TorchVision's
+                `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V2
 
 
 class EfficientNet_B2_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/rwightman/pytorch-image-models/
         url="https://download.pytorch.org/models/efficientnet_b2_rwightman-bcdf34b7.pth",
         transforms=partial(
             ImageClassification, crop_size=288, resize_size=288, interpolation=InterpolationMode.BICUBIC
@@ -511,13 +522,15 @@ class EfficientNet_B2_Weights(WeightsEnum):
                 "acc@1": 80.608,
                 "acc@5": 95.310,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_B3_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/rwightman/pytorch-image-models/
         url="https://download.pytorch.org/models/efficientnet_b3_rwightman-cf984f9c.pth",
         transforms=partial(
             ImageClassification, crop_size=300, resize_size=320, interpolation=InterpolationMode.BICUBIC
@@ -529,13 +542,15 @@ class EfficientNet_B3_Weights(WeightsEnum):
                 "acc@1": 82.008,
                 "acc@5": 96.054,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_B4_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/rwightman/pytorch-image-models/
         url="https://download.pytorch.org/models/efficientnet_b4_rwightman-7eb33cd5.pth",
         transforms=partial(
             ImageClassification, crop_size=380, resize_size=384, interpolation=InterpolationMode.BICUBIC
@@ -547,13 +562,15 @@ class EfficientNet_B4_Weights(WeightsEnum):
                 "acc@1": 83.384,
                 "acc@5": 96.594,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_B5_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/
         url="https://download.pytorch.org/models/efficientnet_b5_lukemelas-b6417697.pth",
         transforms=partial(
             ImageClassification, crop_size=456, resize_size=456, interpolation=InterpolationMode.BICUBIC
@@ -565,13 +582,15 @@ class EfficientNet_B5_Weights(WeightsEnum):
                 "acc@1": 83.444,
                 "acc@5": 96.628,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_B6_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/
         url="https://download.pytorch.org/models/efficientnet_b6_lukemelas-c76e70fd.pth",
         transforms=partial(
             ImageClassification, crop_size=528, resize_size=528, interpolation=InterpolationMode.BICUBIC
@@ -583,13 +602,15 @@ class EfficientNet_B6_Weights(WeightsEnum):
                 "acc@1": 84.008,
                 "acc@5": 96.916,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_B7_Weights(WeightsEnum):
     IMAGENET1K_V1 = Weights(
+        # Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/
         url="https://download.pytorch.org/models/efficientnet_b7_lukemelas-dcc49843.pth",
         transforms=partial(
             ImageClassification, crop_size=600, resize_size=600, interpolation=InterpolationMode.BICUBIC
@@ -601,6 +622,7 @@ class EfficientNet_B7_Weights(WeightsEnum):
                 "acc@1": 84.122,
                 "acc@5": 96.908,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
@@ -622,6 +644,11 @@ class EfficientNet_V2_S_Weights(WeightsEnum):
                 "acc@1": 84.228,
                 "acc@5": 96.878,
             },
+            "_docs": """
+                These weights improve upon the results of the original paper by using a modified version of TorchVision's
+                `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V1
@@ -643,12 +670,18 @@ class EfficientNet_V2_M_Weights(WeightsEnum):
                 "acc@1": 85.112,
                 "acc@5": 97.156,
             },
+            "_docs": """
+                These weights improve upon the results of the original paper by using a modified version of TorchVision's
+                `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V1
 
 
 class EfficientNet_V2_L_Weights(WeightsEnum):
+    # Weights ported from https://github.com/google/automl/tree/master/efficientnetv2
     IMAGENET1K_V1 = Weights(
         url="https://download.pytorch.org/models/efficientnet_v2_l-59c71312.pth",
         transforms=partial(
@@ -666,6 +699,7 @@ class EfficientNet_V2_L_Weights(WeightsEnum):
                 "acc@1": 85.808,
                 "acc@5": 97.788,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
@@ -1036,13 +1070,11 @@ def efficientnet_v2_l(
 
 model_urls = _ModelURLs(
     {
-        # Weights ported from https://github.com/rwightman/pytorch-image-models/
         "efficientnet_b0": EfficientNet_B0_Weights.IMAGENET1K_V1.url,
         "efficientnet_b1": EfficientNet_B1_Weights.IMAGENET1K_V1.url,
         "efficientnet_b2": EfficientNet_B2_Weights.IMAGENET1K_V1.url,
         "efficientnet_b3": EfficientNet_B3_Weights.IMAGENET1K_V1.url,
         "efficientnet_b4": EfficientNet_B4_Weights.IMAGENET1K_V1.url,
-        # Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/
         "efficientnet_b5": EfficientNet_B5_Weights.IMAGENET1K_V1.url,
         "efficientnet_b6": EfficientNet_B6_Weights.IMAGENET1K_V1.url,
         "efficientnet_b7": EfficientNet_B7_Weights.IMAGENET1K_V1.url,

torchvision/models/googlenet.py

@@ -288,6 +288,7 @@ class GoogLeNet_Weights(WeightsEnum):
                 "acc@1": 69.778,
                 "acc@5": 89.530,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1

torchvision/models/inception.py

@@ -420,6 +420,7 @@ class Inception_V3_Weights(WeightsEnum):
                 "acc@1": 77.294,
                 "acc@5": 93.450,
             },
+            "_docs": """These weights are ported from the original paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1

torchvision/models/mnasnet.py

@@ -229,6 +229,7 @@ class MNASNet0_5_Weights(WeightsEnum):
                 "acc@1": 67.734,
                 "acc@5": 87.490,
             },
+            "_docs": """These weights reproduce closely the results of the paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
@@ -246,6 +247,10 @@ class MNASNet0_75_Weights(WeightsEnum):
                 "acc@1": 71.180,
                 "acc@5": 90.496,
             },
+            "_docs": """
+                These weights were trained from scratch by using TorchVision's `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V1
@@ -262,6 +267,7 @@ class MNASNet1_0_Weights(WeightsEnum):
                 "acc@1": 73.456,
                 "acc@5": 91.510,
             },
+            "_docs": """These weights reproduce closely the results of the paper.""",
         },
     )
     DEFAULT = IMAGENET1K_V1
@@ -279,6 +285,10 @@ class MNASNet1_3_Weights(WeightsEnum):
                 "acc@1": 76.506,
                 "acc@5": 93.522,
             },
+            "_docs": """
+                These weights were trained from scratch by using TorchVision's `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V1

torchvision/models/mobilenetv2.py

@@ -212,6 +212,7 @@ class MobileNet_V2_Weights(WeightsEnum):
                 "acc@1": 71.878,
                 "acc@5": 90.286,
             },
+            "_docs": """These weights reproduce closely the results of the paper using a simple training recipe.""",
         },
     )
     IMAGENET1K_V2 = Weights(
@@ -224,6 +225,11 @@ class MobileNet_V2_Weights(WeightsEnum):
                 "acc@1": 72.154,
                 "acc@5": 90.822,
             },
+            "_docs": """
+                These weights improve upon the results of the original paper by using a modified version of TorchVision's
+                `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V2

torchvision/models/mobilenetv3.py

@@ -321,6 +321,7 @@ class MobileNet_V3_Large_Weights(WeightsEnum):
                 "acc@1": 74.042,
                 "acc@5": 91.340,
             },
+            "_docs": """These weights were trained from scratch by using a simple training recipe.""",
         },
     )
     IMAGENET1K_V2 = Weights(
@@ -334,6 +335,11 @@ class MobileNet_V3_Large_Weights(WeightsEnum):
                 "acc@1": 75.274,
                 "acc@5": 92.566,
             },
+            "_docs": """
+                These weights improve marginally upon the results of the original paper by using a modified version of
+                TorchVision's `new training recipe
+                <https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/>`_.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V2
@@ -351,6 +357,9 @@ class MobileNet_V3_Small_Weights(WeightsEnum):
                 "acc@1": 67.668,
                 "acc@5": 87.402,
             },
+            "_docs": """
+                These weights improve upon the results of the original paper by using a simple training recipe.
+            """,
         },
     )
     DEFAULT = IMAGENET1K_V1