torchvision

This library is part of the PyTorch project. PyTorch is an open source machine learning framework.

Features described in this documentation are classified by release status:

Stable: These features will be maintained long-term and there should generally be no major performance limitations or gaps in documentation. We also expect to maintain backwards compatibility (although breaking changes can happen and notice will be given one release ahead of time).

Beta: Features are tagged as Beta because the API may change based on user feedback, because the performance needs to improve, or because coverage across operators is not yet complete. For Beta features, we are committing to seeing the feature through to the Stable classification. We are not, however, committing to backwards compatibility.

Prototype: These features are typically not available as part of binary distributions like PyPI or Conda, except sometimes behind run-time flags, and are at an early stage for feedback and testing.

The torchvision package consists of popular datasets, model architectures, and common image transformations for computer vision.

Package Reference

• Transforming and augmenting images
  • Scriptable transforms
  • Compositions of transforms
  • Transforms on PIL Image and torch.*Tensor
  • Transforms on PIL Image only
  • Transforms on torch.*Tensor only
  • Conversion Transforms
  • Generic Transforms
  • Automatic Augmentation Transforms
  • Functional Transforms
• Models and pre-trained weights
  • Classification
  • Semantic Segmentation
  • Object Detection, Instance Segmentation and Person Keypoint Detection
  • Video classification
  • Optical flow
• Datasets
  • Built-in datasets
  • Base classes for custom datasets
• Utils
  • draw_bounding_boxes
  • draw_segmentation_masks
  • draw_keypoints
  • flow_to_image
  • make_grid
  • save_image
• Operators
  • batched_nms
  • box_area
  • box_convert
  • box_iou
  • clip_boxes_to_image
  • deform_conv2d
  • generalized_box_iou
  • generalized_box_iou_loss
  • masks_to_boxes
  • nms
  • ps_roi_align
  • ps_roi_pool
  • remove_small_boxes
  • roi_align
  • roi_pool
  • sigmoid_focal_loss
  • stochastic_depth
  • RoIAlign
  • PSRoIAlign
  • RoIPool
  • PSRoIPool
  • DeformConv2d
  • MultiScaleRoIAlign
  • FeaturePyramidNetwork
  • StochasticDepth
  • FrozenBatchNorm2d
  • SqueezeExcitation
• Reading/Writing images and videos
  • Video
  • Fine-grained video API
  • Image
• Feature extraction for model inspection
  • API Reference

Examples and training references

• Example gallery
• Training references

torchvision.get_image_backend()

Gets the name of the package used to load images

torchvision.get_video_backend()

Returns the currently active video backend used to decode videos.

Returns

Name of the video backend. one of {‘pyav’, ‘video_reader’}.

Return type

str

torchvision.set_image_backend(backend)

Specifies the package used to load images.

Parameters

backend (string) – Name of the image backend. one of {‘PIL’, ‘accimage’}. The accimage package uses the Intel IPP library. It is generally faster than PIL, but does not support as many operations.

torchvision.set_video_backend(backend)

Specifies the package used to decode videos.

Parameters

backend (string) – Name of the video backend. one of {‘pyav’, ‘video_reader’}. The pyav package uses the 3rd party PyAv library. It is a Pythonic binding for the FFmpeg libraries. The video_reader package includes a native C++ implementation on top of FFMPEG libraries, and a python API of TorchScript custom operator. It generally decodes faster than pyav, but is perhaps less robust.

Note

Building with FFMPEG is disabled by default in the latest main. If you want to use the ‘video_reader’ backend, please compile torchvision from source.

PyTorch Libraries

• PyTorch
• torchaudio
• torchtext
• torchvision
• TorchElastic
• TorchServe
• PyTorch on XLA Devices

Indices

• Index