Big doc revamp - simplify and improve the subpackage pages (#4783)

5b611444 · Nicolas Hug · GitHub · 57b3b42c · 5b611444 · 5b611444
Unverified Commit 5b611444 authored Nov 01, 2021 by Nicolas Hug Committed by GitHub Nov 01, 2021
17 changed files
--- a/.gitignore
+++ b/.gitignore
@@ -14,6 +14,7 @@ docs/build
 # sphinx-gallery
 docs/source/auto_examples/
 docs/source/gen_modules/
+docs/source/generated/
 # pytorch-sphinx-theme gets installed here
 docs/src

--- a/docs/Makefile
+++ b/docs/Makefile
@@ -32,6 +32,7 @@ clean:
 	rm -rf $(BUILDDIR)/*
 	rm -rf $(SOURCEDIR)/auto_examples/  # sphinx-gallery
 	rm -rf $(SOURCEDIR)/gen_modules/  # sphinx-gallery
+	rm -rf $(SOURCEDIR)/generated/  # autosummary
 .PHONY: help Makefile docset

--- a/docs/source/_templates/class.rst
+++ b/docs/source/_templates/class.rst
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: {{ module }}
+{{ name | underline}}
+.. autoclass:: {{ name }}
+    :members:
--- a/docs/source/_templates/class_dataset.rst
+++ b/docs/source/_templates/class_dataset.rst
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: {{ module }}
+{{ name | underline}}
+.. autoclass:: {{ name }}
+    :members:
+        __getitem__,
+        {% if "category_name" in methods %} category_name {% endif %}
+    :special-members:
--- a/docs/source/_templates/function.rst
+++ b/docs/source/_templates/function.rst
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: {{ module }}
+{{ name | underline}}
+.. autofunction:: {{ name }}
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -144,6 +144,9 @@ html_css_files = [
 htmlhelp_basename = "PyTorchdoc"
+autosummary_generate = True
 # -- Options for LaTeX output ---------------------------------------------
 latex_elements = {
    # The paper size ('letterpaper' or 'a4paper').

--- a/docs/source/datasets.rst
+++ b/docs/source/datasets.rst
-torchvision.datasets
+Datasets
-====================
+========
+Torchvision provides many built-in datasets in the ``torchvision.datasets``
+module, as well as utility classes for building your own datasets.
+Built-in datasets
+~~~~~~~~~~~~~~~~~
 All datasets are subclasses of :class:`torch.utils.data.Dataset`
 i.e, they have ``__getitem__`` and ``__len__`` methods implemented.
@@ -19,263 +25,58 @@ All the datasets have almost similar API. They all have two common arguments:
 ``transform`` and  ``target_transform`` to transform the input and target respectively.
 You can also create your own datasets using the provided :ref:`base classes <base_classes_datasets>`.
-Caltech
-~~~~~~~
-.. autoclass:: Caltech101
-  :members: __getitem__
-  :special-members:
-.. autoclass:: Caltech256
-  :members: __getitem__
-  :special-members:
-CelebA
-~~~~~~
-.. autoclass:: CelebA
-  :members: __getitem__
-  :special-members:
-CIFAR
-~~~~~
-.. autoclass:: CIFAR10
-  :members: __getitem__
-  :special-members:
-.. autoclass:: CIFAR100
-Cityscapes
-~~~~~~~~~~
-.. note ::
-    Requires Cityscape to be downloaded.
-.. autoclass:: Cityscapes
-  :members: __getitem__
-  :special-members:
-COCO
-~~~~
-.. note ::
-    These require the `COCO API to be installed`_
-.. _COCO API to be installed: https://github.com/pdollar/coco/tree/master/PythonAPI
-Captions
-^^^^^^^^
-.. autoclass:: CocoCaptions
-  :members: __getitem__
-  :special-members:
-Detection
-^^^^^^^^^
-.. autoclass:: CocoDetection
-  :members: __getitem__
-  :special-members:
-EMNIST
-~~~~~~
-.. autoclass:: EMNIST
-FakeData
-~~~~~~~~
-.. autoclass:: FakeData
-Fashion-MNIST
-~~~~~~~~~~~~~
-.. autoclass:: FashionMNIST
-Flickr
-~~~~~~
-.. autoclass:: Flickr8k
-  :members: __getitem__
-  :special-members:
-.. autoclass:: Flickr30k
-  :members: __getitem__
-  :special-members:
-HMDB51
-~~~~~~~
-.. autoclass:: HMDB51
-  :members: __getitem__
-  :special-members:
-ImageNet
-~~~~~~~~~~~
-.. autoclass:: ImageNet
-.. note ::
-    This requires `scipy` to be installed
-iNaturalist
-~~~~~~~~~~~
-.. autoclass:: INaturalist
-  :members: __getitem__, category_name
-Kinetics-400
-~~~~~~~~~~~~
-.. autoclass:: Kinetics400
-  :members: __getitem__
-  :special-members:
-KITTI
-~~~~~~~~~
-.. autoclass:: Kitti
-  :members: __getitem__
-  :special-members:
-KMNIST
-~~~~~~~~~~~~~
-.. autoclass:: KMNIST
-LFW
-~~~~~
-.. autoclass:: LFWPeople
-  :members: __getitem__
-  :special-members:
-.. autoclass:: LFWPairs
-  :members: __getitem__
-  :special-members:
-LSUN
-~~~~
-.. autoclass:: LSUN
-  :members: __getitem__
-  :special-members:
-MNIST
-~~~~~
-.. autoclass:: MNIST
-Omniglot
-~~~~~~~~
-.. autoclass:: Omniglot
-PhotoTour
-~~~~~~~~~
-.. autoclass:: PhotoTour
-  :members: __getitem__
-  :special-members:
-Places365
-~~~~~~~~~
-.. autoclass:: Places365
-  :members: __getitem__
-  :special-members:
-QMNIST
-~~~~~~
-.. autoclass:: QMNIST
-SBD
-~~~~~~
-.. autoclass:: SBDataset
-  :members: __getitem__
-  :special-members:
-SBU
-~~~
-.. autoclass:: SBU
-  :members: __getitem__
-  :special-members:
-SEMEION
-~~~~~~~
-.. autoclass:: SEMEION
-  :members: __getitem__
-  :special-members:
-STL10
-~~~~~
-.. autoclass:: STL10
-  :members: __getitem__
-  :special-members:
-SVHN
-~~~~~
-.. autoclass:: SVHN
-  :members: __getitem__
-  :special-members:
-UCF101
-~~~~~~~
-.. autoclass:: UCF101
-  :members: __getitem__
-  :special-members:
-USPS
-~~~~~
-.. autoclass:: USPS
-  :members: __getitem__
-  :special-members:
-VOC
-~~~~~~
-.. autoclass:: VOCSegmentation
-  :members: __getitem__
-  :special-members:
-.. autoclass:: VOCDetection
-  :members: __getitem__
-  :special-members:
-WIDERFace
-~~~~~~~~~
-.. autoclass:: WIDERFace
-  :members: __getitem__
-  :special-members:
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+    Caltech101
+    Caltech256
+    CelebA
+    CIFAR10
+    CIFAR100
+    Cityscapes
+    CocoCaptions
+    CocoDetection
+    EMNIST
+    FakeData
+    FashionMNIST
+    Flickr8k
+    Flickr30k
+    HMDB51
+    ImageNet
+    INaturalist
+    Kinetics400
+    Kitti
+    KMNIST
+    LFWPeople
+    LFWPairs
+    LSUN
+    MNIST
+    Omniglot
+    PhotoTour
+    Places365
+    QMNIST
+    SBDataset
+    SBU
+    SEMEION
+    STL10
+    SVHN
+    UCF101
+    USPS
+    VOCSegmentation
+    VOCDetection
+    WIDERFace
 .. _base_classes_datasets:
 Base classes for custom datasets
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-.. autoclass:: DatasetFolder
+.. autosummary::
-  :members: __getitem__, find_classes, make_dataset
+    :toctree: generated/
-  :special-members:
+    :template: class.rst
-.. autoclass:: ImageFolder
-  :members: __getitem__
-  :special-members:
-.. autoclass:: VisionDataset
+    DatasetFolder
-  :members: __getitem__
+    ImageFolder
-  :special-members:
+    VisionDataset
--- a/docs/source/docutils.conf
+++ b/docs/source/docutils.conf
+[html writers]
+table_style: colwidths-auto  # Necessary for the table generated by autosummary to look decent
--- a/docs/source/feature_extraction.rst
+++ b/docs/source/feature_extraction.rst
-torchvision.models.feature_extraction
+Feature extraction for model inspection
-=====================================
+=======================================
 .. currentmodule:: torchvision.models.feature_extraction
-Feature extraction utilities let us tap into our models to access intermediate
+The ``torchvision.models.feature_extraction`` package contains
+feature extraction utilities that let us tap into our models to access intermediate
 transformations of our inputs. This could be useful for a variety of
 applications in computer vision. Just a few examples are:
@@ -157,6 +158,9 @@ Here is an example of how we might extract features for MaskRCNN:
 API Reference
 -------------
-.. autofunction:: create_feature_extractor
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
-.. autofunction:: get_graph_node_names
+    create_feature_extractor
+    get_graph_node_names
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -32,11 +32,11 @@ architectures, and common image transformations for computer vision.
   :caption: Package Reference
   datasets
-   io
+   transforms
   models
   feature_extraction
   ops
-   transforms
+   io
   utils
 .. toctree::

--- a/docs/source/io.rst
+++ b/docs/source/io.rst
-torchvision.io
+Reading/Writing images and videos
-==============
+=================================
 .. currentmodule:: torchvision.io
@@ -10,11 +10,13 @@ images.
 Video
 -----
-.. autofunction:: read_video
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
-.. autofunction:: read_video_timestamps
+    read_video
+    read_video_timestamps
-.. autofunction:: write_video
+    write_video
 Fine-grained video API
@@ -24,8 +26,11 @@ In addition to the :mod:`read_video` function, we provide a high-performance
 lower-level API for more fine-grained control compared to the :mod:`read_video` function.
 It does all this whilst fully supporting torchscript.
-.. autoclass:: VideoReader
+.. autosummary::
-    :members: __next__, get_metadata, set_current_stream, seek
+    :toctree: generated/
+    :template: class.rst
+    VideoReader
 Example of inspecting a video:
@@ -59,24 +64,23 @@ Example of inspecting a video:
 Image
 -----
-.. autoclass:: ImageReadMode
+.. autosummary::
+    :toctree: generated/
-.. autofunction:: read_image
+    :template: class.rst
-.. autofunction:: decode_image
+    ImageReadMode
-.. autofunction:: encode_jpeg
+.. autosummary::
+    :toctree: generated/
-.. autofunction:: decode_jpeg
+    :template: function.rst
-.. autofunction:: write_jpeg
+    read_image
+    decode_image
-.. autofunction:: encode_png
+    encode_jpeg
+    decode_jpeg
-.. autofunction:: decode_png
+    write_jpeg
+    encode_png
-.. autofunction:: write_png
+    decode_png
+    write_png
-.. autofunction:: read_file
+    read_file
+    write_file
-.. autofunction:: write_file
--- a/docs/source/models.rst
+++ b/docs/source/models.rst
 .. _models:
-torchvision.models
+Models and pre-trained weights
-##################
+##############################
-The models subpackage contains definitions of models for addressing
+The ``torchvision.models`` subpackage contains definitions of models for addressing
 different tasks, including: image classification, pixelwise semantic
 segmentation, object detection, instance segmentation, person
 keypoint detection and video classification.
@@ -256,48 +256,72 @@ regnet_y_32gf                     80.878          95.340
 Alexnet
 -------
-.. autofunction:: alexnet
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    alexnet
 VGG
 ---
-.. autofunction:: vgg11
+.. autosummary::
-.. autofunction:: vgg11_bn
+    :toctree: generated/
-.. autofunction:: vgg13
+    :template: function.rst
-.. autofunction:: vgg13_bn
-.. autofunction:: vgg16
+    vgg11
-.. autofunction:: vgg16_bn
+    vgg11_bn
-.. autofunction:: vgg19
+    vgg13
-.. autofunction:: vgg19_bn
+    vgg13_bn
+    vgg16
+    vgg16_bn
+    vgg19
+    vgg19_bn
 ResNet
 ------
-.. autofunction:: resnet18
+.. autosummary::
-.. autofunction:: resnet34
+    :toctree: generated/
-.. autofunction:: resnet50
+    :template: function.rst
-.. autofunction:: resnet101
-.. autofunction:: resnet152
+    resnet18
+    resnet34
+    resnet50
+    resnet101
+    resnet152
 SqueezeNet
 ----------
-.. autofunction:: squeezenet1_0
+.. autosummary::
-.. autofunction:: squeezenet1_1
+    :toctree: generated/
+    :template: function.rst
+    squeezenet1_0
+    squeezenet1_1
 DenseNet
 ---------
-.. autofunction:: densenet121
+.. autosummary::
-.. autofunction:: densenet169
+    :toctree: generated/
-.. autofunction:: densenet161
+    :template: function.rst
-.. autofunction:: densenet201
+    densenet121
+    densenet169
+    densenet161
+    densenet201
 Inception v3
 ------------
-.. autofunction:: inception_v3
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    inception_v3
 .. note ::
    This requires `scipy` to be installed
@@ -306,7 +330,11 @@ Inception v3
 GoogLeNet
 ------------
-.. autofunction:: googlenet
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    googlenet
 .. note ::
    This requires `scipy` to be installed
@@ -315,71 +343,103 @@ GoogLeNet
 ShuffleNet v2
 -------------
-.. autofunction:: shufflenet_v2_x0_5
+.. autosummary::
-.. autofunction:: shufflenet_v2_x1_0
+    :toctree: generated/
-.. autofunction:: shufflenet_v2_x1_5
+    :template: function.rst
-.. autofunction:: shufflenet_v2_x2_0
+    shufflenet_v2_x0_5
+    shufflenet_v2_x1_0
+    shufflenet_v2_x1_5
+    shufflenet_v2_x2_0
 MobileNet v2
 -------------
-.. autofunction:: mobilenet_v2
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    mobilenet_v2
 MobileNet v3
 -------------
-.. autofunction:: mobilenet_v3_large
+.. autosummary::
-.. autofunction:: mobilenet_v3_small
+    :toctree: generated/
+    :template: function.rst
+    mobilenet_v3_large
+    mobilenet_v3_small
 ResNext
 -------
-.. autofunction:: resnext50_32x4d
+.. autosummary::
-.. autofunction:: resnext101_32x8d
+    :toctree: generated/
+    :template: function.rst
+    resnext50_32x4d
+    resnext101_32x8d
 Wide ResNet
 -----------
-.. autofunction:: wide_resnet50_2
+.. autosummary::
-.. autofunction:: wide_resnet101_2
+    :toctree: generated/
+    :template: function.rst
+    wide_resnet50_2
+    wide_resnet101_2
 MNASNet
 --------
-.. autofunction:: mnasnet0_5
+.. autosummary::
-.. autofunction:: mnasnet0_75
+    :toctree: generated/
-.. autofunction:: mnasnet1_0
+    :template: function.rst
-.. autofunction:: mnasnet1_3
+    mnasnet0_5
+    mnasnet0_75
+    mnasnet1_0
+    mnasnet1_3
 EfficientNet
 ------------
-.. autofunction:: efficientnet_b0
+.. autosummary::
-.. autofunction:: efficientnet_b1
+    :toctree: generated/
-.. autofunction:: efficientnet_b2
+    :template: function.rst
-.. autofunction:: efficientnet_b3
-.. autofunction:: efficientnet_b4
+    efficientnet_b0
-.. autofunction:: efficientnet_b5
+    efficientnet_b1
-.. autofunction:: efficientnet_b6
+    efficientnet_b2
-.. autofunction:: efficientnet_b7
+    efficientnet_b3
+    efficientnet_b4
+    efficientnet_b5
+    efficientnet_b6
+    efficientnet_b7
 RegNet
 ------------
-.. autofunction:: regnet_y_400mf
+.. autosummary::
-.. autofunction:: regnet_y_800mf
+    :toctree: generated/
-.. autofunction:: regnet_y_1_6gf
+    :template: function.rst
-.. autofunction:: regnet_y_3_2gf
-.. autofunction:: regnet_y_8gf
+    regnet_y_400mf
-.. autofunction:: regnet_y_16gf
+    regnet_y_800mf
-.. autofunction:: regnet_y_32gf
+    regnet_y_1_6gf
-.. autofunction:: regnet_x_400mf
+    regnet_y_3_2gf
-.. autofunction:: regnet_x_800mf
+    regnet_y_8gf
-.. autofunction:: regnet_x_1_6gf
+    regnet_y_16gf
-.. autofunction:: regnet_x_3_2gf
+    regnet_y_32gf
-.. autofunction:: regnet_x_8gf
+    regnet_x_400mf
-.. autofunction:: regnet_x_16gf
+    regnet_x_800mf
-.. autofunction:: regnet_x_32gf
+    regnet_x_1_6gf
+    regnet_x_3_2gf
+    regnet_x_8gf
+    regnet_x_16gf
+    regnet_x_32gf
 Quantized Models
 ----------------
@@ -473,22 +533,34 @@ LR-ASPP MobileNetV3-Large         57.9           91.2
 Fully Convolutional Networks
 ----------------------------
-.. autofunction:: torchvision.models.segmentation.fcn_resnet50
+.. autosummary::
-.. autofunction:: torchvision.models.segmentation.fcn_resnet101
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.segmentation.fcn_resnet50
+    torchvision.models.segmentation.fcn_resnet101
 DeepLabV3
 ---------
-.. autofunction:: torchvision.models.segmentation.deeplabv3_resnet50
+.. autosummary::
-.. autofunction:: torchvision.models.segmentation.deeplabv3_resnet101
+    :toctree: generated/
-.. autofunction:: torchvision.models.segmentation.deeplabv3_mobilenet_v3_large
+    :template: function.rst
+    torchvision.models.segmentation.deeplabv3_resnet50
+    torchvision.models.segmentation.deeplabv3_resnet101
+    torchvision.models.segmentation.deeplabv3_mobilenet_v3_large
 LR-ASPP
 -------
-.. autofunction:: torchvision.models.segmentation.lraspp_mobilenet_v3_large
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.segmentation.lraspp_mobilenet_v3_large
 .. _object_det_inst_seg_pers_keypoint_det:
@@ -615,39 +687,63 @@ Keypoint R-CNN ResNet-50 FPN            0.3789               0.1242
 Faster R-CNN
 ------------
-.. autofunction:: torchvision.models.detection.fasterrcnn_resnet50_fpn
+.. autosummary::
-.. autofunction:: torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn
+    :toctree: generated/
-.. autofunction:: torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn
+    :template: function.rst
+    torchvision.models.detection.fasterrcnn_resnet50_fpn
+    torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn
+    torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn
 RetinaNet
 ---------
-.. autofunction:: torchvision.models.detection.retinanet_resnet50_fpn
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.detection.retinanet_resnet50_fpn
 SSD
 ---
-.. autofunction:: torchvision.models.detection.ssd300_vgg16
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.detection.ssd300_vgg16
 SSDlite
 -------
-.. autofunction:: torchvision.models.detection.ssdlite320_mobilenet_v3_large
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.detection.ssdlite320_mobilenet_v3_large
 Mask R-CNN
 ----------
-.. autofunction:: torchvision.models.detection.maskrcnn_resnet50_fpn
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.detection.maskrcnn_resnet50_fpn
 Keypoint R-CNN
 --------------
-.. autofunction:: torchvision.models.detection.keypointrcnn_resnet50_fpn
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.detection.keypointrcnn_resnet50_fpn
 Video classification
@@ -686,14 +782,26 @@ ResNet (2+1)D                     57.50           78.81
 ResNet 3D
 ----------
-.. autofunction:: torchvision.models.video.r3d_18
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.video.r3d_18
 ResNet Mixed Convolution
 ------------------------
-.. autofunction:: torchvision.models.video.mc3_18
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.video.mc3_18
 ResNet (2+1)D
 -------------
-.. autofunction:: torchvision.models.video.r2plus1d_18
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+    torchvision.models.video.r2plus1d_18
--- a/docs/source/ops.rst
+++ b/docs/source/ops.rst
 .. _ops:
-torchvision.ops
+Operators
-===============
+=========
 .. currentmodule:: torchvision.ops
@@ -10,29 +10,36 @@ torchvision.ops
 .. note::
  All operators have native support for TorchScript.
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
-.. autofunction:: batched_nms
+    batched_nms
-.. autofunction:: box_area
+    box_area
-.. autofunction:: box_convert
+    box_convert
-.. autofunction:: box_iou
+    box_iou
-.. autofunction:: clip_boxes_to_image
+    clip_boxes_to_image
-.. autofunction:: deform_conv2d
+    deform_conv2d
-.. autofunction:: generalized_box_iou
+    generalized_box_iou
-.. autofunction:: masks_to_boxes
+    masks_to_boxes
-.. autofunction:: nms
+    nms
-.. autofunction:: ps_roi_align
+    ps_roi_align
-.. autofunction:: ps_roi_pool
+    ps_roi_pool
-.. autofunction:: remove_small_boxes
+    remove_small_boxes
-.. autofunction:: roi_align
+    roi_align
-.. autofunction:: roi_pool
+    roi_pool
-.. autofunction:: sigmoid_focal_loss
+    sigmoid_focal_loss
-.. autofunction:: stochastic_depth
+    stochastic_depth
-.. autoclass:: RoIAlign
+.. autosummary::
-.. autoclass:: PSRoIAlign
+    :toctree: generated/
-.. autoclass:: RoIPool
+    :template: class.rst
-.. autoclass:: PSRoIPool
-.. autoclass:: DeformConv2d
+    RoIAlign
-.. autoclass:: MultiScaleRoIAlign
+    PSRoIAlign
-.. autoclass:: FeaturePyramidNetwork
+    RoIPool
-.. autoclass:: StochasticDepth
+    PSRoIPool
+    DeformConv2d
+    MultiScaleRoIAlign
+    FeaturePyramidNetwork
+    StochasticDepth
--- a/docs/source/transforms.rst
+++ b/docs/source/transforms.rst
 .. _transforms:
-torchvision.transforms
+Transforming and augmenting images
-======================
+==================================
 .. currentmodule:: torchvision.transforms
-Transforms are common image transformations. They can be chained together using :class:`Compose`.
+Transforms are common image transformations available in the
+``torchvision.transforms`` module. They can be chained together using
+:class:`Compose`.
 Most transform classes have a function equivalent: :ref:`functional
 transforms <functional_transforms>` give fine-grained control over the
 transformations.
@@ -90,131 +92,94 @@ For any custom transformations to be used with ``torch.jit.script``, they should
 Compositions of transforms
 --------------------------
-.. autoclass:: Compose
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+    Compose
 Transforms on PIL Image and torch.\*Tensor
 ------------------------------------------
-.. autoclass:: CenterCrop
+.. autosummary::
-    :members:
+    :toctree: generated/
+    :template: class.rst
-.. autoclass:: ColorJitter
-    :members:
+    CenterCrop
+    ColorJitter
-.. autoclass:: FiveCrop
+    FiveCrop
-    :members:
+    Grayscale
+    Pad
-.. autoclass:: Grayscale
+    RandomAffine
-    :members:
+    RandomApply
+    RandomCrop
-.. autoclass:: Pad
+    RandomGrayscale
-    :members:
+    RandomHorizontalFlip
+    RandomPerspective
-.. autoclass:: RandomAffine
+    RandomResizedCrop
-    :members:
+    RandomRotation
+    RandomSizedCrop
-.. autoclass:: RandomApply
+    RandomVerticalFlip
+    Resize
-.. autoclass:: RandomCrop
+    TenCrop
-    :members:
+    GaussianBlur
+    RandomInvert
-.. autoclass:: RandomGrayscale
+    RandomPosterize
-    :members:
+    RandomSolarize
+    RandomAdjustSharpness
-.. autoclass:: RandomHorizontalFlip
+    RandomAutocontrast
-    :members:
+    RandomEqualize
-.. autoclass:: RandomPerspective
-    :members:
-.. autoclass:: RandomResizedCrop
-    :members:
-.. autoclass:: RandomRotation
-    :members:
-.. autoclass:: RandomSizedCrop
-    :members:
-.. autoclass:: RandomVerticalFlip
-    :members:
-.. autoclass:: Resize
-    :members:
-.. autoclass:: Scale
-    :members:
-.. autoclass:: TenCrop
-    :members:
-.. autoclass:: GaussianBlur
-    :members:
-.. autoclass:: RandomInvert
-    :members:
-.. autoclass:: RandomPosterize
-    :members:
-.. autoclass:: RandomSolarize
-    :members:
-.. autoclass:: RandomAdjustSharpness
-    :members:
-.. autoclass:: RandomAutocontrast
-    :members:
-.. autoclass:: RandomEqualize
-    :members:
 .. _transforms_pil_only:
 Transforms on PIL Image only
 ----------------------------
-.. autoclass:: RandomChoice
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
-.. autoclass:: RandomOrder
+    RandomChoice
+    RandomOrder
 .. _transforms_tensor_only:
 Transforms on torch.\*Tensor only
 ---------------------------------
-.. autoclass:: LinearTransformation
+.. autosummary::
-    :members:
+    :toctree: generated/
+    :template: class.rst
-.. autoclass:: Normalize
-    :members:
-.. autoclass:: RandomErasing
+    LinearTransformation
-    :members:
+    Normalize
+    RandomErasing
-.. autoclass:: ConvertImageDtype
+    ConvertImageDtype
 .. _conversion_transforms:
 Conversion Transforms
 ---------------------
-.. autoclass:: ToPILImage
+.. autosummary::
-    :members:
+    :toctree: generated/
+    :template: class.rst
-.. autoclass:: ToTensor
-    :members:
-.. autoclass:: PILToTensor
+    ToPILImage
-    :members:
+    ToTensor
+    PILToTensor
 Generic Transforms
 ------------------
-.. autoclass:: Lambda
+.. autosummary::
-    :members:
+    :toctree: generated/
+    :template: class.rst
+    Lambda
 Automatic Augmentation Transforms
@@ -226,27 +191,22 @@ ImageNet policies provide significant improvements when applied to other dataset
 In TorchVision we implemented 3 policies learned on the following datasets: ImageNet, CIFAR10 and SVHN.
 The new transform can be used standalone or mixed-and-matched with existing transforms:
-.. autoclass:: AutoAugmentPolicy
+.. autosummary::
-    :members:
+    :toctree: generated/
+    :template: class.rst
-.. autoclass:: AutoAugment
-    :members:
-`RandAugment <https://arxiv.org/abs/1909.13719>`_ is a simple high-performing Data Augmentation technique which improves the accuracy of Image Classification models.
+    AutoAugmentPolicy
+    AutoAugment
-.. autoclass:: RandAugment
+    RandAugment
-    :members:
+    TrivialAugmentWide
-`TrivialAugmentWide <https://arxiv.org/abs/2103.10158>`_ is a dataset-independent data-augmentation technique which improves the accuracy of Image Classification models.
-.. autoclass:: TrivialAugmentWide
-    :members:
 .. _functional_transforms:
 Functional Transforms
 ---------------------
+.. currentmodule:: torchvision.transforms.functional
 Functional transforms give you fine-grained control of the transformation pipeline.
 As opposed to the transformations above, functional transforms don't contain a random number
 generator for their parameters.
@@ -291,5 +251,41 @@ you can use a functional transform to build transform classes with custom behavi
    rotation_transform = MyRotationTransform(angles=[-30, -15, 0, 15, 30])
-.. automodule:: torchvision.transforms.functional
+.. autosummary::
-    :members:
+    :toctree: generated/
+    :template: function.rst
+    adjust_brightness
+    adjust_contrast
+    adjust_gamma
+    adjust_hue
+    adjust_saturation
+    adjust_sharpness
+    affine
+    autocontrast
+    center_crop
+    convert_image_dtype
+    crop
+    equalize
+    erase
+    five_crop
+    gaussian_blur
+    get_image_num_channels
+    get_image_size
+    hflip
+    invert
+    normalize
+    pad
+    perspective
+    pil_to_tensor
+    posterize
+    resize
+    resized_crop
+    rgb_to_grayscale
+    rotate
+    solarize
+    ten_crop
+    to_grayscale
+    to_pil_image
+    to_tensor
+    vflip
--- a/docs/source/utils.rst
+++ b/docs/source/utils.rst
 .. _utils:
-torchvision.utils
+Utils
-=================
+=====
-.. currentmodule:: torchvision.utils
+The ``torchvision.utils`` module contains various utilities, mostly :ref:`for
+vizualization <sphx_glr_auto_examples_plot_visualization_utils.py>`. 
-.. autofunction:: make_grid
-.. autofunction:: save_image
+.. currentmodule:: torchvision.utils
-.. autofunction:: draw_bounding_boxes
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
-.. autofunction:: draw_segmentation_masks
+    draw_bounding_boxes
+    draw_segmentation_masks
+    make_grid
+    save_image
--- a/torchvision/datasets/coco.py
+++ b/torchvision/datasets/coco.py
@@ -10,6 +10,8 @@ from .vision import VisionDataset
 class CocoDetection(VisionDataset):
    """`MS Coco Detection <https://cocodataset.org/#detection-2016>`_ Dataset.
+    It requires the `COCO API to be installed <https://github.com/pdollar/coco/tree/master/PythonAPI>`_.
    Args:
        root (string): Root directory where images are downloaded to.
        annFile (string): Path to json annotation file.
@@ -59,6 +61,8 @@ class CocoDetection(VisionDataset):
 class CocoCaptions(CocoDetection):
    """`MS Coco Captions <https://cocodataset.org/#captions-2015>`_ Dataset.
+    It requires the `COCO API to be installed <https://github.com/pdollar/coco/tree/master/PythonAPI>`_.
    Args:
        root (string): Root directory where images are downloaded to.
        annFile (string): Path to json annotation file.

--- a/torchvision/ops/deform_conv.py
+++ b/torchvision/ops/deform_conv.py
@@ -161,12 +161,10 @@ class DeformConv2d(nn.Module):
        """
        Args:
            input (Tensor[batch_size, in_channels, in_height, in_width]): input tensor
-            offset (Tensor[batch_size, 2 * offset_groups * kernel_height * kernel_width,
+            offset (Tensor[batch_size, 2 * offset_groups * kernel_height * kernel_width, out_height, out_width]):
-                out_height, out_width]): offsets to be applied for each position in the
+                offsets to be applied for each position in the convolution kernel.
-                convolution kernel.
+            mask (Tensor[batch_size, offset_groups * kernel_height * kernel_width, out_height, out_width]):
-            mask (Tensor[batch_size, offset_groups * kernel_height * kernel_width,
+                masks to be applied for each position in the convolution kernel.
-                out_height, out_width]): masks to be applied for each position in the
-                convolution kernel.
        """
        return deform_conv2d(
            input,