docs: Added MaskRCNN to the new doc (#5900)

* docs: Added MaskRCNN to new doc

* docs: Updated docstring

* ufmt
Co-authored-by: FG Fernandez <–26927750+frgfm@users.noreply.github.com>
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

docs/source/models/mask_rcnn.rst

+Mask R-CNN
+==========
+
+.. currentmodule:: torchvision.models.detection
+
+The Mask R-CNN model is based on the `Mask R-CNN <https://arxiv.org/abs/1703.06870>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a Mask R-CNN model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.mask_rcnn.MaskRCNN`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    maskrcnn_resnet50_fpn
+    maskrcnn_resnet50_fpn_v2

docs/source/models_new.rst

@@ -94,6 +94,7 @@ weights:
    :maxdepth: 1
 
    models/retinanet
+   models/mask_rcnn
 
 Table of all available detection weights
 ----------------------------------------

torchvision/models/detection/mask_rcnn.py

@@ -403,10 +403,8 @@ def maskrcnn_resnet50_fpn(
     trainable_backbone_layers: Optional[int] = None,
     **kwargs: Any,
 ) -> MaskRCNN:
-    """
-    Constructs a Mask R-CNN model with a ResNet-50-FPN backbone.
-
-    Reference: `"Mask R-CNN" <https://arxiv.org/abs/1703.06870>`_.
+    """Mask R-CNN model with a ResNet-50-FPN backbone from the `Mask R-CNN
+    <https://arxiv.org/abs/1703.06870>`_ paper.
 
     The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each
     image, and should be in ``0-1`` range. Different images can have different sizes.
@@ -451,13 +449,26 @@ def maskrcnn_resnet50_fpn(
         >>> torch.onnx.export(model, x, "mask_rcnn.onnx", opset_version = 11)
 
     Args:
-        weights (MaskRCNN_ResNet50_FPN_Weights, optional): The pretrained weights for the model
-        progress (bool): If True, displays a progress bar of the download to stderr
+        weights (:class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights`, optional): The
+            pretrained weights to use. See
+            :class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights` below for
+            more details, and possible values. By default, no pre-trained
+            weights are used.
+        progress (bool, optional): If True, displays a progress bar of the
+            download to stderr. Default is True.
         num_classes (int, optional): number of output classes of the model (including the background)
-        weights_backbone (ResNet50_Weights, optional): The pretrained weights for the backbone
-        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from final block.
-            Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If ``None`` is
-            passed (the default) this value is set to 3.
+        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The
+            pretrained weights for the backbone.
+        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from
+            final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+            trainable. If ``None`` is passed (the default) this value is set to 3.
+        **kwargs: parameters passed to the ``torchvision.models.detection.mask_rcnn.MaskRCNN``
+            base class. Please refer to the `source code
+            <https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py>`_
+            for more details about this class.
+
+    .. autoclass:: torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights
+        :members:
     """
     weights = MaskRCNN_ResNet50_FPN_Weights.verify(weights)
     weights_backbone = ResNet50_Weights.verify(weights_backbone)
@@ -493,22 +504,32 @@ def maskrcnn_resnet50_fpn_v2(
     trainable_backbone_layers: Optional[int] = None,
     **kwargs: Any,
 ) -> MaskRCNN:
-    """
-    Constructs an improved MaskRCNN model with a ResNet-50-FPN backbone.
-
-    Reference: `"Benchmarking Detection Transfer Learning with Vision Transformers"
-    <https://arxiv.org/abs/2111.11429>`_.
+    """Improved Mask R-CNN model with a ResNet-50-FPN backbone from the `Benchmarking Detection Transfer
+    Learning with Vision Transformers <https://arxiv.org/abs/2111.11429>`_ paper.
 
     :func:`~torchvision.models.detection.maskrcnn_resnet50_fpn` for more details.
 
     Args:
-        weights (MaskRCNN_ResNet50_FPN_V2_Weights, optional): The pretrained weights for the model
-        progress (bool): If True, displays a progress bar of the download to stderr
+        weights (:class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights`, optional): The
+            pretrained weights to use. See
+            :class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights` below for
+            more details, and possible values. By default, no pre-trained
+            weights are used.
+        progress (bool, optional): If True, displays a progress bar of the
+            download to stderr. Default is True.
         num_classes (int, optional): number of output classes of the model (including the background)
-        weights_backbone (ResNet50_Weights, optional): The pretrained weights for the backbone
-        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from final block.
-            Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If ``None`` is
-            passed (the default) this value is set to 3.
+        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The
+            pretrained weights for the backbone.
+        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from
+            final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+            trainable. If ``None`` is passed (the default) this value is set to 3.
+        **kwargs: parameters passed to the ``torchvision.models.detection.mask_rcnn.MaskRCNN``
+            base class. Please refer to the `source code
+            <https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py>`_
+            for more details about this class.
+
+    .. autoclass:: torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights
+        :members:
     """
     weights = MaskRCNN_ResNet50_FPN_V2_Weights.verify(weights)
     weights_backbone = ResNet50_Weights.verify(weights_backbone)