Oneflow-Inc /oneflow
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.
★ 9,434⑂ 1,013👁 361⑃ 9,602● 1,067▤ 34
+0
Daily
9,434
stars
1,013
forks
361
watchers
9,602
prs
1,067
issues
34
releases
README
OneFlow
OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient. With OneFlow, it is easy to:
- program a model with PyTorch-like API
- scale a model to n-dimensional-parallel execution with the Global Tensor
- accelerate/deploy a model with the Graph Compiler.
Latest News
- Version 1.0.0 is out!
- Full changelog
Publication
- OneFlow: Redesign the Distributed Deep Learning Framework from Scratch
System Requirements
General
- Linux
- Python 3.7, 3.8, 3.9, 3.10, 3.11
CUDA
- CUDA arch 60 or above
- CUDA Toolkit version 10.0 or above
- Nvidia driver version 440.33 or above
OneFlow will work on a minimum supported driver, and any driver beyond. For more information, please refer to CUDA compatibility documentation.
Install
Preinstall docker image
docker pull oneflowinc/oneflow:nightly-cuda11.8
Pip Install
- (Highly recommended) Upgrade pip
python3 -m pip install --upgrade pip --user
- To install latest stable release of OneFlow with CUDA support:
bash
python3 -m pip install oneflow
- To install nightly release of OneFlow with CPU-only support:
bash
python3 -m pip install --pre oneflow -f https://oneflow-staging.oss-cn-beijing.aliyuncs.com/branch/master/cpu
- To install nightly release of OneFlow with CUDA support:
bash
python3 -m pip install --pre oneflow -f https://oneflow-staging.oss-cn-beijing.aliyuncs.com/branch/master/cu118
If you are in China, you could run this to have pip download packages from domestic mirror of pypi:
python3 -m pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
For more information on this, please refer to pypi 镜像使用帮助
Install from Source
Clone Source Code
- Option 1: Clone source code from GitHub
bash
git clone https://github.com/Oneflow-Inc/oneflow.git
- Option 2: Download from Aliyun(Only available in China)
bash
curl https://oneflow-public.oss-cn-beijing.aliyuncs.com/oneflow-src.zip -o oneflow-src.zip
unzip oneflow-src.zip
Build OneFlow
- Install dependencies
apt install -y libopenblas-dev nasm g++ gcc python3-pip cmake autoconf libtool
These dependencies are preinstalled in offical conda environment and docker image, you can use the offical conda environment here or use the docker image by:
bash
docker pull oneflowinc/manylinux2014_x86_64_cuda11.2
- In the root directory of OneFlow source code, run:
mkdir build
cd build
- Config the project, inside build directory:
- If you are in China
config for CPU-only like this:
cmake .. -C ../cmake/caches/cn/cpu.cmake
config for CUDA like this:
cmake .. -C ../cmake/caches/cn/cuda.cmake -DCMAKE_CUDA_ARCHITECTURES=80 -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda -DCUDNN_ROOT_DIR=/usr/local/cudnn
- If you are not in China
config for CPU-only like this:
cmake .. -C ../cmake/caches/international/cpu.cmake
config for CUDA likTop issues
#9075带有 BatchNorm2d 的模型在开启 amp 和 grad acc 时会报错29 comments · 2022-09-09#3516F0828 13:01:44.340785 7127 job_completer.cpp:71] Check failed: mutable_consumer == nullptr 14 comments · 2020-08-28#7020Consistent tensor indexing12 comments · 2021-12-13#3784数据集加载12 comments · 2020-11-09#8881为算子添加性能测试(profile)11 comments · 2022-08-09#3882Check failed: job().job_conf().train_conf().has_model_update_conf()11 comments · 2020-11-21#9398RuntimeError: This package is a placeholder. Please install oneflow following the instructions10 comments · 2022-11-09#8910NeuS网络训练速度慢10 comments · 2022-08-12
Discussions
#10319Multi graphics card training error0 comments · 2023-08-23#10222Anything from oneflow for new mov_to_mov0 comments · 2023-05-03#9849OneEmbedding的推理方案是什么?1 comments · 2023-02-09#9848[ROCm] ROCm support?2 comments · 2023-02-08#9450需求:在macOs/ios/windows下可用oneflow推理加速0 comments · 2022-11-20#9326CUTLASS Integration1 comments · 2022-10-27#9272学习了下persistent table相关实现,但发现没有compaction机制,一直顺序追加写磁盘空间不会爆嘛1 comments · 2022-10-19#9058Users Questions from Wechat1 comments · 2022-09-06
Latest release
Version 1.0.0 v1.0.0 2024-03-11Version 1.0.0 OneFlow v1.0.0 release note OneFlow v1.0.0 came out, welcome to install the new version for a better experience. - Highlights - New Features - Improvements - Changes and Fixes - Performance Highlights This version update includes 447 commits and the following highlights: - Released a new interface compile_from_torch. This interface, while sharing the parameter memory, converts a PyTorch Module instance into a OneFlow Module instance. It supports direct Eager execution or conversion into a static graph nn.Graph, further accelerating the process using MLIR c