diff --git a/README-zh-Hans.md b/README-zh-Hans.md index 8ff25c64f..8a3bee5ec 100644 --- a/README-zh-Hans.md +++ b/README-zh-Hans.md @@ -274,19 +274,26 @@ Colossal-AI 为您提供了一系列并行组件。我们的目标是让您的
(返回顶端)
## 安装 -> Colossal-AI 目前仅支持Linux操作系统,没有在其他操作系统如Windows和macOS进行测试。 -> -> 环境要求: PyTorch 1.10 ~ 1.12 (更新版本正在兼容中), Python >= 3.7, CUDA >= 11.0。如果你遇到安装问题,可以向本项目 [反馈](https://github.com/hpcaitech/ColossalAI/issues/new/choose)。 + +环境要求: + +- PyTorch >= 1.11 (PyTorch 2.x 正在适配中) +- Python >= 3.7 +- CUDA >= 11.0 + +如果你遇到安装问题,可以向本项目 [反馈](https://github.com/hpcaitech/ColossalAI/issues/new/choose)。 ### 从PyPI安装 -您可以用下面的命令直接从PyPI上下载并安装Colossal-AI。我们默认不会安装PyTorch扩展包 +您可以用下面的命令直接从PyPI上下载并安装Colossal-AI。我们默认不会安装PyTorch扩展包。 ```bash pip install colossalai ``` +**注:目前只支持Linux。** + 但是,如果你想在安装时就直接构建PyTorch扩展,您可以设置环境变量`CUDA_EXT=1`. ```bash diff --git a/README.md b/README.md index 10d59e34c..3115192d6 100644 --- a/README.md +++ b/README.md @@ -276,9 +276,13 @@ Acceleration of [AlphaFold Protein Structure](https://alphafold.ebi.ac.uk/) ## Installation -> Colossal-AI currently only supports the Linux operating system and has not been tested on other OS such as Windows and macOS. -> -> Environment Requirement: PyTorch 1.10 ~ 1.12 (WIP higher version), Python >= 3.7, CUDA >= 11.0. If you encounter any problem about installation, you may want to raise an [issue](https://github.com/hpcaitech/ColossalAI/issues/new/choose) in this repository. + +Requirements: +- PyTorch >= 1.11 (PyTorch 2.x in progress) +- Python >= 3.7 +- CUDA >= 11.0 + +If you encounter any problem about installation, you may want to raise an [issue](https://github.com/hpcaitech/ColossalAI/issues/new/choose) in this repository. ### Install from PyPI @@ -288,6 +292,8 @@ You can easily install Colossal-AI with the following command. **By default, we pip install colossalai ``` +**Note: only Linux is supported for now.** + However, if you want to build the PyTorch extensions during installation, you can set `CUDA_EXT=1`. ```bash diff --git a/docs/source/en/get_started/installation.md b/docs/source/en/get_started/installation.md index da30ab4e5..c1c34b411 100644 --- a/docs/source/en/get_started/installation.md +++ b/docs/source/en/get_started/installation.md @@ -1,9 +1,13 @@ # Setup -> Colossal-AI currently only supports the Linux operating system and has not been tested on other OS such as Windows and macOS. -> -> Environment Requirement: PyTorch 1.10 ~ 1.12 (WIP higher version), Python >= 3.7, CUDA >= 11.0. If you encounter any problem about installation, you may want to raise an [issue](https://github.com/hpcaitech/ColossalAI/issues/new/choose) in this repository. + +Requirements: +- PyTorch >= 1.11 (PyTorch 2.x in progress) +- Python >= 3.7 +- CUDA >= 11.0 + +If you encounter any problem about installation, you may want to raise an [issue](https://github.com/hpcaitech/ColossalAI/issues/new/choose) in this repository. ## Download From PyPI @@ -14,6 +18,8 @@ You can install Colossal-AI with pip install colossalai ``` +**Note: only Linux is supported for now** + If you want to build PyTorch extensions during installation, you can use the command below. Otherwise, the PyTorch extensions will be built during runtime. ```shell diff --git a/docs/source/zh-Hans/get_started/installation.md b/docs/source/zh-Hans/get_started/installation.md index 2ceb0231f..fb79fc676 100755 --- a/docs/source/zh-Hans/get_started/installation.md +++ b/docs/source/zh-Hans/get_started/installation.md @@ -1,7 +1,12 @@ # 安装 -> Colossal-AI 目前仅支持Linux操作系统,没有在其他操作系统如Windows和macOS进行测试 -> -> 环境要求: PyTorch 1.10 ~ 1.12 (更新版本正在兼容中), Python >= 3.7, CUDA >= 11.0。如果你遇到安装问题,可以向本项目 [反馈](https://github.com/hpcaitech/ColossalAI/issues/new/choose)。 + +环境要求: + +- PyTorch >= 1.11 (PyTorch 2.x 正在适配中) +- Python >= 3.7 +- CUDA >= 11.0 + +如果你遇到安装问题,可以向本项目 [反馈](https://github.com/hpcaitech/ColossalAI/issues/new/choose)。 ## 从PyPI上安装 @@ -11,6 +16,8 @@ pip install colossalai ``` +**注:现在只支持Linux。** + 如果你想同时安装PyTorch扩展的话,可以添加`CUDA_EXT=1`。如果不添加的话,PyTorch扩展会在运行时自动安装。 ```shell diff --git a/setup.py b/setup.py index 6c24cb504..0a66a9008 100644 --- a/setup.py +++ b/setup.py @@ -1,4 +1,5 @@ import os +import sys from datetime import datetime from typing import List @@ -31,6 +32,10 @@ IS_NIGHTLY = int(os.environ.get('NIGHTLY', '0')) == 1 # a variable to store the op builder ext_modules = [] +# we do not support windows currently +if sys.platform == 'win32': + raise RuntimeError("Windows is not supported yet. Please try again within the Windows Subsystem for Linux (WSL).") + # check for CUDA extension dependencies def environment_check_for_cuda_extension_build():