pytorch ORG

!PyTorch Logo

--------------------------------------------------------------------------------

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 at hud.pytorch.org.

<!-- toc -->

• [More About PyTorch](#more-about-pytorch)

- [A GPU-Ready Tensor Library](#a-gpu-ready-tensor-library) - [Dynamic Neural Networks: Tape-Based Autograd](#dynamic-neural-networks-tape-based-autograd) - [Python First](#python-first) - [Imperative Experiences](#imperative-experiences) - [Fast and Lean](#fast-and-lean) - [Extensions Without Pain](#extensions-without-pain)

• [Installation](#installation)

- [Binaries](#binaries) - [NVIDIA Jetson Platforms](#nvidia-jetson-platforms) - [From Source](#from-source) - [Prerequisites](#prerequisites) - [NVIDIA CUDA Support](#nvidia-cuda-support) - [AMD ROCm Support](#amd-rocm-support) - [Intel GPU Support](#intel-gpu-support) - [Get the PyTorch Source](#get-the-pytorch-source) - [Install Dependencies](#install-dependencies) - [Install PyTorch](#install-pytorch) - [Adjust Build Options (Optional)](#adjust-build-options-optional) - [Docker Image](#docker-image) - [Using pre-built images](#using-pre-built-images) - [Building the image yourself](#building-the-image-yourself) - [Building the Documentation](#building-the-documentation) - [Troubleshooting CI Errors](#troubleshooting-ci-errors) - [Building a PDF](#building-a-pdf) - [Previous Versions](#previous-versions)

• [Getting Started](#getting-started)
• [Resources](#resources)
• [Communication](#communication)
• [Releases and Contributing](#releases-and-contributing)
• [The Team](#the-team)
• [License](#license)

<!-- tocstop -->

More About PyTorch

Learn the basics of PyTorch

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

| Component | Description |
| ---- | --- |
| torch | A Tensor library like NumPy, with strong GPU support |
| torch.autograd | A tape-based automatic differentiation library that supports all differentiable Tensor operations in torch |
| torch.jit | A compilation stack (TorchScript) to create serializable and optimizable models from PyTorch code |
| torch.nn | A neural networks library deeply integrated with autograd designed for maximum flexibility |
| torch.multiprocessing | Python multiprocessing, but with magical memory sharing of torch Tensors across processes. Useful for data loading and Hogwild training |
| torch.utils | DataLoader and other utility functions for convenience |

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:

A GPU-Ready Tensor Library

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

!Tensor illustration

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

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

Dynamic Neural Networks: Tape-Based Autograd

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 to
change the way your network behaves arbitrarily with zero lag or overhead. Our inspiration comes
from several research papers on this topic, as well as current and past work such as
torch-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.

!Dynamic graph

Python First

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 use NumPy / SciPy / scikit-learn etc.
You can write your new neural network layers in Python itself, using your favorite libraries
and use packages such as Cython and Numba.
Our goal is to not reinvent the wheel where appropriate.

Imperative Experiences

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.

Fast and Lean

PyTorch has minimal framework overhead. We integrate acceleration libraries
such as Intel MKL and NVIDIA (cuDNN, NCCL) to maximize speed.
At the core, its CPU and GPU Tensor and neural network backends
are 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 that
your deep learning models are maximally memory efficient.
This enables you to train bigger deep learning models than before.

Extensions Without Pain

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

You can write new neural network layers in Python using the torch API
or 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 see a tutorial here and an example here.

Installation

Binaries

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

NVIDIA Jetson Platforms

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

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

From Source

Prerequisites

If you are installing from source, you will need:

• Python 3.10 or later
• A compiler that fully supports C++20, such as clang or gcc (gcc 11.3.0 or newer is required, on Linux)
• Visual Studio or Visual Studio Build Tool (Windows only)
• At least 10 GB of free disk space
• 30-60 minutes for the initial build (subsequent rebuilds are much faster)

\* PyTorch CI uses Visual C++ BuildTools, which come with Visual Studio Enterprise, Professional, or Community Editions. You can also install the build tools from https://visualstudio.microsoft.com/visual-cpp-build-tools/. The build tools *do not* come with Visual Studio Code by default.

An example of environment setup is shown below:

• Linux:

bash
$ source /bin/activate
$ conda create -y -n 
$ conda activate

• Windows:

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

A conda environment is not required. You can also do a PyTorch build in a
standard virtual environment, e.g., created with tools like uv, provided
your system has installed all the necessary dependencies unavailable as pip
packages (e.g., CUDA, MKL.)

NVIDIA CUDA Support

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 v9.0 or above
• Compiler compatible with CUDA

Note: You could refer to the cuDNN 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 variable USE_CUDA=0.
Other potentially useful environment variables may be found in setup.py. If
CUDA is installed in a non-standard location, set PATH so that the nvcc you
want 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 are available here

AMD ROCm Support

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, the ROCM_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 the PYTORCH_ROCM_ARCH environment variable AMD GPU architecture

If you want to disable ROCm support, export the environment variable USE_ROCM=0.
Other potentially useful environment variables may be found in setup.py.

Intel GPU Support

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 variable USE_XPU=0. Other potentially useful environment variables may be found in setup.py.

Get the PyTorch Source

bash
git clone https://github.com/pytorch/pytorch
cd pytorch
# if you are updating an existing checkout
git submodule sync
git submodule update --init --recursive

Install Dependencies

Common

bash
# Run this command from the PyTorch directory after cloning the source code using the “Get the PyTorch Source“ section above
pip install --group dev

On Linux

bash
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 Windows

bash
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=1.51

Install PyTorch

On Linux

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

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

Install PyTorch

bash
# the CMake prefix for conda environment
export CMAKE_PREFIX_PATH="${CONDA_PREFIX:-'$(dirname $(which conda))/../'}:${CMAKE_PREFIX_PATH}"
python -m pip install --no-build-isolation -v -e .

# the CMake prefix for non-conda environment, e.g. Python venv
# call following after activating the venv
export CMAKE_PREFIX_PATH="${VIRTUAL_ENV}:${CMAKE_PREFIX_PATH}"

On macOS

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

On Windows

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

CPU-only builds

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

cmd
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 tweaking CMAKE_INCLUDE_PATH and LIB. The instruction here 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. If ninja.exe is detected in PATH, 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 as
Magma, oneDNN, a.k.a. MKLDNN or DNNL, and Sccache are often needed. Please refer to the

…

Collapse