| commit | 1ad683033b73480b87db753a4ad29f34c0d8a1c8 | [log] [tgz] |
|---|---|---|
| author | Haoci Zhang <haocizhang@meta.com> | Tue Jul 02 07:54:35 2024 +0000 |
| committer | PyTorch MergeBot <pytorchmergebot@users.noreply.github.com> | Tue Jul 02 07:54:38 2024 +0000 |
| tree | 73f823c7ec9fe24946bd19277a0e373f1f5e0d7a | |
| parent | 3e2df3ca9d0a593e09bc94c14bbf2b213413cbf3 [diff] |
Implemented flexible PP schedule (#129597) Enabled some cases to work where num_microbatches % pp_size != 0. Using the flex_pp schedule, we will have num_rounds = max(1, n_microbatches // pp_group_size) and it works as long as n_microbatches % num_rounds is 0. As a few examples, support pp_group_size = 4, n_microbatches = 10. We will have num_rounds = 2 and n_microbatches % 2 is 0. pp_group_size = 4, n_microbatches = 3. We will have num_rounds = 1 and n_microbatches % 1 is 0. Moved over from PiPPy (https://github.com/pytorch/PiPPy/pull/1129) Tested using the config in (1), schedule looks like the following graph: ``` =========== ALL_RANK_ACTIONS =========== Rank 0 Rank 1 Rank 2 Rank 3 Step 00: F0_s0 None None None Step 01: F1_s0 F0_s1 None None Step 02: F2_s0 F1_s1 F0_s2 None Step 03: F3_s0 F2_s1 F1_s2 F0_s3 Step 04: F4_s0 F3_s1 F2_s2 F1_s3 Step 05: F0_s4 F4_s1 F3_s2 F2_s3 Step 06: F1_s4 F0_s5 F4_s2 F3_s3 Step 07: F2_s4 F1_s5 F0_s6 F4_s3 Step 08: F3_s4 F2_s5 F1_s6 F0_s7 Step 09: F4_s4 F3_s5 None B0_s7 Step 10: F5_s0 None F2_s6 F1_s7 Step 11: None None B0_s6 B1_s7 Step 12: None F4_s5 F3_s6 F2_s7 Step 13: None B0_s5 B1_s6 B2_s7 Step 14: F6_s0 F5_s1 F4_s6 F3_s7 Step 15: B0_s4 B1_s5 B2_s6 B3_s7 Step 16: F7_s0 F6_s1 F5_s2 F4_s7 Step 17: B1_s4 B2_s5 B3_s6 B4_s7 Step 18: F8_s0 F7_s1 F6_s2 F5_s3 Step 19: B2_s4 B3_s5 B4_s6 B0_s3 Step 20: F9_s0 F8_s1 F7_s2 F6_s3 Step 21: B3_s4 B4_s5 B0_s2 B1_s3 Step 22: F5_s4 F9_s1 F8_s2 F7_s3 Step 23: B4_s4 B0_s1 B1_s2 B2_s3 Step 24: F6_s4 F5_s5 F9_s2 F8_s3 Step 25: B0_s0 B1_s1 B2_s2 B3_s3 Step 26: F7_s4 F6_s5 F5_s6 F9_s3 Step 27: B1_s0 B2_s1 B3_s2 B4_s3 Step 28: F8_s4 F7_s5 F6_s6 F5_s7 Step 29: B2_s0 B3_s1 B4_s2 B5_s7 Step 30: F9_s4 F8_s5 F7_s6 F6_s7 Step 31: B3_s0 B4_s1 B5_s6 B6_s7 Step 32: None F9_s5 F8_s6 F7_s7 Step 33: B4_s0 B5_s5 B6_s6 B7_s7 Step 34: None None F9_s6 F8_s7 Step 35: B5_s4 B6_s5 B7_s6 B8_s7 Step 36: None None None F9_s7 Step 37: B6_s4 B7_s5 B8_s6 B9_s7 Step 38: None None None None Step 39: B7_s4 B8_s5 B9_s6 B5_s3 Step 40: None None None None Step 41: B8_s4 B9_s5 B5_s2 B6_s3 Step 42: None None None None Step 43: B9_s4 B5_s1 B6_s2 B7_s3 Step 44: None None None None Step 45: B5_s0 B6_s1 B7_s2 B8_s3 Step 46: None None None None Step 47: B6_s0 B7_s1 B8_s2 B9_s3 Step 48: None None None Step 49: B7_s0 B8_s1 B9_s2 Step 50: None None Step 51: B8_s0 B9_s1 Step 52: None ``` Pull Request resolved: https://github.com/pytorch/pytorch/pull/129597 Approved by: https://github.com/H-Huang
PyTorch is a Python package that provides two high-level features:
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.
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:
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 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!
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.
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.
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 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.
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.
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 provided here and the L4T container is published here
They require JetPack 4.2 and above, and @dusty-nv and @ptrblck are maintaining them.
If you are installing from source, you will need:
We highly recommend installing an Anaconda environment. You will get a high-quality BLAS library (MKL) and you get controlled dependency versions regardless of your Linux distro.
If you want to compile with CUDA support, select a supported version of CUDA from our support matrix, then install the following:
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 you are building for NVIDIA's Jetson platforms (Jetson Nano, TX1, TX2, AGX Xavier), Instructions to install PyTorch for Jetson Nano are available here
If you want to compile with ROCm support, install
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.
If you want to compile with Intel GPU support, follow these
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.
Common
conda install cmake ninja # Run this command from the PyTorch directory after cloning the source code using the “Get the PyTorch Source“ section below pip install -r requirements.txt
On Linux
conda install intel::mkl-static intel::mkl-include # CUDA only: Add LAPACK support for the GPU if needed conda install -c pytorch magma-cuda121 # or the magma-cuda* that matches your CUDA version from https://anaconda.org/pytorch/repo # (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 only conda install intel::mkl-static intel::mkl-include # Add these packages if torch.distributed is needed conda install pkg-config libuv
On Windows
conda install intel::mkl-static intel::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
git clone --recursive https://github.com/pytorch/pytorch cd pytorch # if you are updating an existing checkout git submodule sync git submodule update --init --recursive
On Linux
If you would like to compile PyTorch with new C++ ABI enabled, then first run this command:
export _GLIBCXX_USE_CXX11_ABI=1
If you're compiling for AMD ROCm then first run this command:
# Only run this if you're compiling for ROCm python tools/amd_build/build_amd.py
Install PyTorch
export CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"} python setup.py develop
Aside: If you are using Anaconda, you may experience an error caused by the linker:
build/temp.linux-x86_64-3.7/torch/csrc/stub.o: file not recognized: file format not recognized collect2: error: ld returned 1 exit status error: command 'g++' failed with exit status 1This is caused by
ldfrom the Conda environment shadowing the systemld. You should use a newer version of Python that fixes this issue. The recommended Python version is 3.8.1+.
On macOS
python3 setup.py develop
On Windows
Choose Correct Visual Studio Version.
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.
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
conda activate python setup.py develop
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 installation-helper to install them.
You can refer to the build_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`. set CMAKE_INCLUDE_PATH={Your directory}\mkl\include set LIB={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. set CMAKE_GENERATOR_TOOLSET_VERSION=14.27 set DISTUTILS_USE_SDK=1 for /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 compiler set CUDAHOSTCXX=C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\bin\HostX64\x64\cl.exe python setup.py develop
You can adjust the configuration of cmake variables optionally (without building first), by doing the following. For example, adjusting the pre-detected directories for CuDNN or BLAS can be done with such a step.
On Linux
export CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"} python setup.py build --cmake-only ccmake build # or cmake-gui build
On macOS
export CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"} MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py build --cmake-only ccmake build # or cmake-gui build
You can also pull a pre-built docker image from Docker Hub and run with docker v19.03+
docker run --gpus all --rm -ti --ipc=host pytorch/pytorch:latest
Please note that PyTorch uses shared memory to share data between processes, so if torch multiprocessing is used (e.g. for multithreaded data loaders) the default shared memory segment size that container runs with is not enough, and you should increase shared memory size either with --ipc=host or --shm-size command line options to nvidia-docker run.
NOTE: Must be built with a docker version > 18.06
The Dockerfile is supplied to build images with CUDA 11.1 support and cuDNN v8. You can pass PYTHON_VERSION=x.y make variable to specify which Python version is to be used by Miniconda, or leave it unset to use the default.
make -f docker.Makefile # images are tagged as docker.io/${your_docker_username}/pytorch
You can also pass the CMAKE_VARS="..." environment variable to specify additional CMake variables to be passed to CMake during the build. See setup.py for the list of available variables.
make -f docker.Makefile
To build documentation in various formats, you will need Sphinx and the readthedocs theme.
cd docs/ pip install -r requirements.txt
You can then build the documentation by running make <format> from the docs/ folder. Run make to get a list of all available output formats.
If you get a katex error run npm install katex. If it persists, try npm install -g katex
Note: if you installed
nodejswith a different package manager (e.g.,conda) thennpmwill probably install a version ofkatexthat is not compatible with your version ofnodejsand doc builds will fail. A combination of versions that is known to work isnode@6.13.1andkatex@0.13.18. To install the latter withnpmyou can runnpm install -g katex@0.13.18
Installation instructions and binaries for previous PyTorch versions may be found on our website.
Three-pointers to get you started:
Typically, PyTorch has three minor releases a year. Please let us know if you encounter a bug by filing 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 our Contribution page. For more information about PyTorch releases, see Release page.
PyTorch is a community-driven project with several skillful engineers and researchers contributing to it.
PyTorch is currently maintained by Soumith Chintala, Gregory Chanan, Dmytro Dzhulgakov, Edward Yang, and Nikita 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 to hughperkins/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 the LICENSE file.