Repository files navigation

PyTorch is a Python package that provides two high-level features:

• Tensor computation (like NumPy) with strong GPU acceleration
• Deep neural networks built on a tape-based autograd system

You can reuse your favorite Python packages such as NumPy, SciPy, and Cython to extend PyTorch when needed.

Our trunk health (Continuous Integration signals) can be found athud.pytorch.org.

• More About PyTorch
  • A GPU-Ready Tensor Library
  • Dynamic Neural Networks: Tape-Based Autograd
  • Python First
  • Imperative Experiences
  • Fast and Lean
  • Extensions Without Pain
• Installation
  • Binaries
    • NVIDIA Jetson Platforms
  • From Source
    • Prerequisites
      • NVIDIA CUDA Support
      • AMD ROCm Support
      • Intel GPU Support
    • Get the PyTorch Source
    • Install Dependencies
    • Install PyTorch
      • Adjust Build Options (Optional)
  • Docker Image
    • Using pre-built images
    • Building the image yourself
  • Building the Documentation
    • Building a PDF
  • Previous Versions
• Getting Started
• Resources
• Communication
• Releases and Contributing
• The Team
• License

Learn the basics of PyTorch

At a granular level, PyTorch is a library that consists of the following components:

Usually, PyTorch is used either as:

• A replacement for NumPy to use the power of GPUs.
• A deep learning research platform that provides maximum flexibility and speed.

Elaborating Further:

If you use NumPy, then you have used Tensors (a.k.a. ndarray).

PyTorch provides Tensors that can live either on the CPU or the GPU and accelerates thecomputation by a huge amount.

We provide a wide variety of tensor routines to accelerate and fit your scientific computation needssuch as slicing, indexing, mathematical operations, linear algebra, reductions.And they are fast!

PyTorch has a unique way of building neural networks: using and replaying a tape recorder.

Most frameworks such as TensorFlow, Theano, Caffe, and CNTK have a static view of the world.One has to build a neural network and reuse the same structure again and again.Changing the way the network behaves means that one has to start from scratch.

With PyTorch, we use a technique called reverse-mode auto-differentiation, which allows you tochange the way your network behaves arbitrarily with zero lag or overhead. Our inspiration comesfrom several research papers on this topic, as well as current and past work such astorch-autograd,autograd,Chainer, etc.

While this technique is not unique to PyTorch, it's one of the fastest implementations of it to date.You get the best of speed and flexibility for your crazy research.

PyTorch is not a Python binding into a monolithic C++ framework.It is built to be deeply integrated into Python.You can use it naturally like you would useNumPy /SciPy /scikit-learn etc.You can write your new neural network layers in Python itself, using your favorite librariesand use packages such asCython andNumba.Our goal is to not reinvent the wheel where appropriate.

PyTorch is designed to be intuitive, linear in thought, and easy to use.When you execute a line of code, it gets executed. There isn't an asynchronous view of the world.When you drop into a debugger or receive error messages and stack traces, understanding them is straightforward.The stack trace points to exactly where your code was defined.We hope you never spend hours debugging your code because of bad stack traces or asynchronous and opaque execution engines.

PyTorch has minimal framework overhead. We integrate acceleration librariessuch asIntel MKL and NVIDIA (cuDNN,NCCL) to maximize speed.At the core, its CPU and GPU Tensor and neural network backendsare mature and have been tested for years.

Hence, PyTorch is quite fast — whether you run small or large neural networks.

The memory usage in PyTorch is extremely efficient compared to Torch or some of the alternatives.We've written custom memory allocators for the GPU to make sure thatyour deep learning models are maximally memory efficient.This enables you to train bigger deep learning models than before.

Writing new neural network modules, or interfacing with PyTorch's Tensor API was designed to be straightforwardand with minimal abstractions.

You can write new neural network layers in Python using the torch APIor your favorite NumPy-based libraries such as SciPy.

If you want to write your layers in C/C++, we provide a convenient extension API that is efficient and with minimal boilerplate.No wrapper code needs to be written. You can seea tutorial here andan example here.

Commands to install binaries via Conda or pip wheels are on our website:https://pytorch.org/get-started/locally/

Python wheels for NVIDIA's Jetson Nano, Jetson TX1/TX2, Jetson Xavier NX/AGX, and Jetson AGX Orin are providedhere and the L4T container is publishedhere

They require JetPack 4.2 and above, and@dusty-nv and@ptrblck are maintaining them.

If you are installing from source, you will need:

• Python 3.9 or later
• A compiler that fully supports C++17, such as clang or gcc (gcc 9.4.0 or newer is required, on Linux)
• Visual Studio or Visual Studio Build Tool (Windows only)

* PyTorch CI uses Visual C++ BuildTools, which come with Visual Studio Enterprise,Professional, or Community Editions. You can also install the build tools fromhttps://visualstudio.microsoft.com/visual-cpp-build-tools/. The build toolsdo notcome with Visual Studio Code by default.

An example of environment setup is shown below:

• Linux:

$source<CONDA_INSTALL_DIR>/bin/activate$ conda create -y -n<CONDA_NAME>$ conda activate<CONDA_NAME>

• Windows:

$source<CONDA_INSTALL_DIR>\Scripts\activate.bat$ conda create -y -n<CONDA_NAME>$ conda activate<CONDA_NAME>$ call"C:\Program Files\Microsoft Visual Studio\<VERSION>\Community\VC\Auxiliary\Build\vcvarsall.bat" x64

A conda environment is not required. You can also do a PyTorch build in astandard virtual environment, e.g., created with tools likeuv, providedyour system has installed all the necessary dependencies unavailable as pippackages (e.g., CUDA, MKL.)

If you want to compile with CUDA support,select a supported version of CUDA from our support matrix, then install the following:

• NVIDIA CUDA
• NVIDIA cuDNN v8.5 or above
• Compiler compatible with CUDA

Note: You could refer to thecuDNN Support Matrix for cuDNN versions with the various supported CUDA, CUDA driver, and NVIDIA hardware.

If you want to disable CUDA support, export the environment variableUSE_CUDA=0.Other potentially useful environment variables may be found insetup.py. IfCUDA is installed in a non-standard location, set PATH so that the nvcc youwant to use can be found (e.g.,export PATH=/usr/local/cuda-12.8/bin:$PATH).

If you are building for NVIDIA's Jetson platforms (Jetson Nano, TX1, TX2, AGX Xavier), Instructions to install PyTorch for Jetson Nano areavailable here

If you want to compile with ROCm support, install

• AMD ROCm 4.0 and above installation
• ROCm is currently supported only for Linux systems.

By default the build system expects ROCm to be installed in/opt/rocm. If ROCm is installed in a different directory, theROCM_PATH environment variable must be set to the ROCm installation directory. The build system automatically detects the AMD GPU architecture. Optionally, the AMD GPU architecture can be explicitly set with thePYTORCH_ROCM_ARCH environment variableAMD GPU architecture

If you want to disable ROCm support, export the environment variableUSE_ROCM=0.Other potentially useful environment variables may be found insetup.py.

If you want to compile with Intel GPU support, follow these

• PyTorch Prerequisites for Intel GPUs instructions.
• Intel GPU is supported for Linux and Windows.

If you want to disable Intel GPU support, export the environment variableUSE_XPU=0.Other potentially useful environment variables may be found insetup.py.

git clone https://github.com/pytorch/pytorchcd pytorch# if you are updating an existing checkoutgit submodule syncgit submodule update --init --recursive

Common

conda install cmake ninja# Run this command from the PyTorch directory after cloning the source code using the “Get the PyTorch Source“ section belowpip install -r requirements.txt

On Linux

pip install mkl-static mkl-include# CUDA only: Add LAPACK support for the GPU if needed# magma installation: run with active conda environment. specify CUDA version to install.ci/docker/common/install_magma_conda.sh 12.4# (optional) If using torch.compile with inductor/triton, install the matching version of triton# Run from the pytorch directory after cloning# For Intel GPU support, please explicitly `export USE_XPU=1` before running command.make triton

On MacOS

# Add this package on intel x86 processor machines onlypip install mkl-static mkl-include# Add these packages if torch.distributed is neededconda install pkg-config libuv

On Windows

pip install mkl-static mkl-include# Add these packages if torch.distributed is needed.# Distributed package support on Windows is a prototype feature and is subject to changes.conda install -c conda-forge libuv=1.39

On Linux

If you're compiling for AMD ROCm then first run this command:

# Only run this if you're compiling for ROCmpython tools/amd_build/build_amd.py

Install PyTorch

export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"python -m pip install -r requirements-build.txtpython -m pip install --no-build-isolation -v -e.

On macOS

python -m pip install -r requirements-build.txtpython -m pip install --no-build-isolation -v -e.

On Windows

If you want to build legacy python code, please refer toBuilding on legacy code and CUDA

CPU-only builds

In this mode PyTorch computations will run on your CPU, not your GPU.

python -m pip install --no-build-isolation -v -e .

Note on OpenMP: The desired OpenMP implementation is Intel OpenMP (iomp). In order to link against iomp, you'll need to manually download the library and set up the building environment by tweakingCMAKE_INCLUDE_PATH andLIB. The instructionhere is an example for setting up both MKL and Intel OpenMP. Without these configurations for CMake, Microsoft Visual C OpenMP runtime (vcomp) will be used.

CUDA based build

In this mode PyTorch computations will leverage your GPU via CUDA for faster number crunching

NVTX is needed to build Pytorch with CUDA.NVTX is a part of CUDA distributive, where it is called "Nsight Compute". To install it onto an already installed CUDA run CUDA installation once again and check the corresponding checkbox.Make sure that CUDA with Nsight Compute is installed after Visual Studio.

Currently, VS 2017 / 2019, and Ninja are supported as the generator of CMake. Ifninja.exe is detected inPATH, then Ninja will be used as the default generator, otherwise, it will use VS 2017 / 2019.
If Ninja is selected as the generator, the latest MSVC will get selected as the underlying toolchain.

Additional libraries such asMagma,oneDNN, a.k.a. MKLDNN or DNNL, andSccache are often needed. Please refer to theinstallation-helper to install them.

You can refer to thebuild_pytorch.bat script for some other environment variables configurations

cmd:: Set the environment variables after you have downloaded and unzipped the mkl package,:: else CMake would throw an error as `Could NOT find OpenMP`.setCMAKE_INCLUDE_PATH={Your directory}\mkl\includesetLIB={Your directory}\mkl\lib;%LIB%:: Read the content in the previous section carefully before you proceed.:: [Optional] If you want to override the underlying toolset used by Ninja and Visual Studio with CUDA, please run the following script block.:: "Visual Studio 2019 Developer Command Prompt" will be run automatically.:: Make sure you have CMake >= 3.12 before you do this when you use the Visual Studio generator.setCMAKE_GENERATOR_TOOLSET_VERSION=14.27setDISTUTILS_USE_SDK=1for /f"usebackq tokens=*"%i in (`"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -version [15^,17^) -products * -latest -property installationPath`) do call "%i\VC\Auxiliary\Build\vcvarsall.bat" x64 -vcvars_ver=%CMAKE_GENERATOR_TOOLSET_VERSION%:: [Optional] If you want to override the CUDA host compilersetCUDAHOSTCXX=C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\bin\HostX64\x64\cl.exepython -m pip install --no-build-isolation -v -e .

Intel GPU builds

In this mode PyTorch with Intel GPU support will be built.

Please make surethe common prerequisites as well asthe prerequisites for Intel GPU are properly installed and the environment variables are configured prior to starting the build. For build tool support,Visual Studio 2022 is required.

Then PyTorch can be built with the command:

:: CMD Commands::: Set the CMAKE_PREFIX_PATH to help find corresponding packages:: %CONDA_PREFIX% only works after `conda activate custom_env`ifdefined CMAKE_PREFIX_PATH (set"CMAKE_PREFIX_PATH=%CONDA_PREFIX%\Library;%CMAKE_PREFIX_PATH%")else (set"CMAKE_PREFIX_PATH=%CONDA_PREFIX%\Library")python -m pip install --no-build-isolation -v -e .

You can adjust the configuration of cmake variables optionally (without building first), by doingthe following. For example, adjusting the pre-detected directories for CuDNN or BLAS can be donewith such a step.

On Linux

export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"CMAKE_ONLY=1 python setup.py buildccmake build# or cmake-gui build

On macOS

export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ CMAKE_ONLY=1 python setup.py buildccmake build# or cmake-gui build

You can also pull a pre-built docker image from Docker Hub and run with docker v19.03+

docker run --gpus all --rm -ti --ipc=host pytorch/pytorch:latest

Please note that PyTorch uses shared memory to share data between processes, so if torch multiprocessing is used (e.g.for multithreaded data loaders) the default shared memory segment size that container runs with is not enough, and youshould increase shared memory size either with--ipc=host or--shm-size command line options tonvidia-docker run.

NOTE: Must be built with a docker version > 18.06

TheDockerfile is supplied to build images with CUDA 11.1 support and cuDNN v8.You can passPYTHON_VERSION=x.y make variable to specify which Python version is to be used by Miniconda, or leave itunset to use the default.

make -f docker.Makefile# images are tagged as docker.io/${your_docker_username}/pytorch

You can also pass theCMAKE_VARS="..." environment variable to specify additional CMake variables to be passed to CMake during the build.Seesetup.py for the list of available variables.

make -f docker.Makefile

To build documentation in various formats, you will needSphinxand the pytorch_sphinx_theme2.

Before you build the documentation locally, ensuretorch isinstalled in your environment. For small fixes, you can install thenightly version as described inGetting Started.

For more complex fixes, such as adding a new module and docstrings forthe new module, you might need to install torchfrom source.SeeDocstring Guidelinesfor docstring conventions.

cd docs/pip install -r requirements.txtmake htmlmake serve

Runmake to get a list of all available output formats.

If you get a katex error runnpm install katex. If it persists, trynpm install -g katex

pip install 'numpy<2'

When you make changes to the dependencies run by CI, edit the.ci/docker/requirements-docs.txt file.

To compile a PDF of all PyTorch documentation, ensure you havetexlive and LaTeX installed. On macOS, you can install them using:

brew install --cask mactex

To create the PDF:

1. Run:

   make latexpdf

   This will generate the necessary files in thebuild/latex directory.

2. Navigate to this directory and execute:

   make LATEXOPTS="-interaction=nonstopmode"

   This will produce apytorch.pdf with the desired content. Run thiscommand one more time so that it generates the correct tableof contents and index.

Installation instructions and binaries for previous PyTorch versions may be foundonour website.

Three pointers to get you started:

• Tutorials: get you started with understanding and using PyTorch
• Examples: easy to understand PyTorch code across all domains
• The API Reference
• Glossary

• PyTorch.org
• PyTorch Tutorials
• PyTorch Examples
• PyTorch Models
• Intro to Deep Learning with PyTorch from Udacity
• Intro to Machine Learning with PyTorch from Udacity
• Deep Neural Networks with PyTorch from Coursera
• PyTorch Twitter
• PyTorch Blog
• PyTorch YouTube

• Forums: Discuss implementations, research, etc.https://discuss.pytorch.org
• GitHub Issues: Bug reports, feature requests, install issues, RFCs, thoughts, etc.
• Slack: ThePyTorch Slack hosts a primary audience of moderate to experienced PyTorch users and developers for general chat, online discussions, collaboration, etc. If you are a beginner looking for help, the primary medium isPyTorch Forums. If you need a slack invite, please fill this form:https://goo.gl/forms/PP1AGvNHpSaJP8to1
• Newsletter: No-noise, a one-way email newsletter with important announcements about PyTorch. You can sign-up here:https://eepurl.com/cbG0rv
• Facebook Page: Important announcements about PyTorch.https://www.facebook.com/pytorch
• For brand guidelines, please visit our website atpytorch.org

Typically, PyTorch has three minor releases a year. Please let us know if you encounter a bug byfiling an issue.

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 to the core, please first open an issue and discuss the feature with us.Sending a PR without discussion might end up resulting in a rejected PR because we might be taking the core in a different direction than you might be aware of.

To learn more about making a contribution to Pytorch, please see ourContribution page. For more information about PyTorch releases, seeRelease page.

PyTorch is a community-driven project with several skillful engineers and researchers contributing to it.

PyTorch is currently maintained bySoumith Chintala,Gregory Chanan,Dmytro Dzhulgakov,Edward Yang, andNikita Shulga with major contributions coming from hundreds of talented individuals in various forms and means.A non-exhaustive but growing list needs to mention:Trevor Killeen,Sasank Chilamkurthy,Sergey Zagoruyko,Adam Lerer,Francisco Massa,Alykhan Tejani,Luca Antiga,Alban Desmaison,Andreas Koepf,James Bradbury,Zeming Lin,Yuandong Tian,Guillaume Lample,Marat Dukhan,Natalia Gimelshein,Christian Sarofeen,Martin Raison,Edward Yang,Zachary Devito.

Note: This project is unrelated tohughperkins/pytorch with the same name. Hugh is a valuable contributor to the Torch community and has helped with many things Torch and PyTorch.

PyTorch has a BSD-style license, as found in theLICENSE file.