torchvision

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

Installation

We recommend Anaconda as Python package management system. Please refer to pytorch.org for the detail of PyTorch (torch) installation. The following is the corresponding torchvision versions and supported Python versions.

torch	torchvision	python
master / nightly	master / nightly	>=3.6
1.5.0	0.6.0	>=3.5
1.4.0	0.5.0	==2.7, >=3.5, <=3.8
1.3.1	0.4.2	==2.7, >=3.5, <=3.7
1.3.0	0.4.1	==2.7, >=3.5, <=3.7
1.2.0	0.4.0	==2.7, >=3.5, <=3.7
1.1.0	0.3.0	==2.7, >=3.5, <=3.7
<=1.0.1	0.2.2	==2.7, >=3.5, <=3.7

Anaconda:

conda install torchvision -c pytorch

pip:

pip install torchvision

From source:

python setup.py install
# or, for OSX
# MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py install

By default, GPU support is built if CUDA is found and torch.cuda.is_available() is true. It's possible to force building GPU support by setting FORCE_CUDA=1 environment variable, which is useful when building a docker image.

Image Backend

Torchvision currently supports the following image backends:

• Pillow (default)
• Pillow-SIMD - a much faster drop-in replacement for Pillow with SIMD. If installed will be used as the default.
• accimage - if installed can be activated by calling torchvision.set_image_backend('accimage')
• libpng - can be installed via conda conda install libpng or any of the package managers for debian-based and RHEL-based Linux distributions.
• libturbojpeg - blazing speed, fast JPEG image loading. Can be installed from conda-forge conda install libjpeg-turbo -c conda-forge.

Notes: libpng and libturbojpeg must be available at compilation time in order to be available. Also, most linux distributions distinguish between libturbojpeg and libjpeg-turbo, where the former should be installed instead of the latter one.

C++ API

TorchVision also offers a C++ API that contains C++ equivalent of python models.

Installation From source:

mkdir build
cd build
# Add -DWITH_CUDA=on support for the CUDA if needed
cmake ..
make
make install

Once installed, the library can be accessed in cmake (after properly configuring CMAKE_PREFIX_PATH) via the TorchVision::TorchVision target:

find_package(TorchVision REQUIRED)
target_link_libraries(my-target PUBLIC TorchVision::TorchVision)

The TorchVision package will also automatically look for the Torch package and add it as a dependency to my-target, so make sure that it is also available to cmake via the CMAKE_PREFIX_PATH.

For an example setup, take a look at examples/cpp/hello_world.

Documentation

You can find the API documentation on the pytorch website: http://pytorch.org/docs/master/torchvision/

Contributing

We appreciate all contributions. If you are planning to contribute back bug-fixes, please do so without any further discussion. If you plan to contribute new features, utility functions or extensions, please first open an issue and discuss the feature with us.

Disclaimer on Datasets

This is a utility library that downloads and prepares public datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have license to use the dataset. It is your responsibility to determine whether you have permission to use the dataset under the dataset's license.

If you're a dataset owner and wish to update any part of it (description, citation, etc.), or do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thanks for your contribution to the ML community!