Merge branch 'main' into prototype-datasets-inheritance

Author: pmeier

.github/workflows/bandit.yml

@@ -65,7 +65,7 @@ def __init__(
         image_ids = set_ids[self._splits_map[self._split]].tolist()
 
         labels = loadmat(self._base_folder / self._file_dict["label"][0], squeeze_me=True)
-        image_id_to_label = dict(enumerate(labels["labels"].tolist(), 1))
+        image_id_to_label = dict(enumerate((labels["labels"] - 1).tolist(), 1))
 
         self._labels = []
         self._image_files = []

.github/workflows/codeql.yml

@@ -393,7 +393,18 @@ class FasterRCNN_ResNet50_FPN_Weights(WeightsEnum):
 
 
 class FasterRCNN_ResNet50_FPN_V2_Weights(WeightsEnum):
-    pass
+    COCO_V1 = Weights(
+        url="https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_v2_coco-dd69338a.pth",
+        transforms=ObjectDetection,
+        meta={
+            **_COMMON_META,
+            "publication_year": 2021,
+            "num_params": 43712278,
+            "recipe": "https://github.com/pytorch/vision/pull/5763",
+            "map": 46.7,
+        },
+    )
+    DEFAULT = COCO_V1
 
 
 class FasterRCNN_MobileNet_V3_Large_FPN_Weights(WeightsEnum):

torchvision/datasets/flowers102.py

@@ -349,17 +349,22 @@ def __init__(self, in_channels, dim_reduced, num_classes):
             #     nn.init.constant_(param, 0)
 
 
+_COMMON_META = {
+    "task": "image_object_detection",
+    "architecture": "MaskRCNN",
+    "categories": _COCO_CATEGORIES,
+    "interpolation": InterpolationMode.BILINEAR,
+}
+
+
 class MaskRCNN_ResNet50_FPN_Weights(WeightsEnum):
     COCO_V1 = Weights(
         url="https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth",
         transforms=ObjectDetection,
         meta={
-            "task": "image_object_detection",
-            "architecture": "MaskRCNN",
+            **_COMMON_META,
             "publication_year": 2017,
             "num_params": 44401393,
-            "categories": _COCO_CATEGORIES,
-            "interpolation": InterpolationMode.BILINEAR,
             "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn",
             "map": 37.9,
             "map_mask": 34.6,
@@ -369,7 +374,19 @@ class MaskRCNN_ResNet50_FPN_Weights(WeightsEnum):
 
 
 class MaskRCNN_ResNet50_FPN_V2_Weights(WeightsEnum):
-    pass
+    COCO_V1 = Weights(
+        url="https://download.pytorch.org/models/maskrcnn_resnet50_fpn_v2_coco-73cbd019.pth",
+        transforms=ObjectDetection,
+        meta={
+            **_COMMON_META,
+            "publication_year": 2021,
+            "num_params": 46359409,
+            "recipe": "https://github.com/pytorch/vision/pull/5773",
+            "map": 47.4,
+            "map_mask": 41.8,
+        },
+    )
+    DEFAULT = COCO_V1
 
 
 @handle_legacy_interface(

torchvision/models/detection/faster_rcnn.py

@@ -672,17 +672,22 @@ def forward(self, images, targets=None):
         return self.eager_outputs(losses, detections)
 
 
+_COMMON_META = {
+    "task": "image_object_detection",
+    "architecture": "RetinaNet",
+    "categories": _COCO_CATEGORIES,
+    "interpolation": InterpolationMode.BILINEAR,
+}
+
+
 class RetinaNet_ResNet50_FPN_Weights(WeightsEnum):
     COCO_V1 = Weights(
         url="https://download.pytorch.org/models/retinanet_resnet50_fpn_coco-eeacb38b.pth",
         transforms=ObjectDetection,
         meta={
-            "task": "image_object_detection",
-            "architecture": "RetinaNet",
+            **_COMMON_META,
             "publication_year": 2017,
             "num_params": 34014999,
-            "categories": _COCO_CATEGORIES,
-            "interpolation": InterpolationMode.BILINEAR,
             "recipe": "https://github.com/pytorch/vision/tree/main/references/detection#retinanet",
             "map": 36.4,
         },
@@ -691,7 +696,18 @@ class RetinaNet_ResNet50_FPN_Weights(WeightsEnum):
 
 
 class RetinaNet_ResNet50_FPN_V2_Weights(WeightsEnum):
-    pass
+    COCO_V1 = Weights(
+        url="https://download.pytorch.org/models/retinanet_resnet50_fpn_v2_coco-5905b1c5.pth",
+        transforms=ObjectDetection,
+        meta={
+            **_COMMON_META,
+            "publication_year": 2019,
+            "num_params": 38198935,
+            "recipe": "https://github.com/pytorch/vision/pull/5756",
+            "map": 41.5,
+        },
+    )
+    DEFAULT = COCO_V1
 
 
 @handle_legacy_interface(