diff --git a/README-zh-Hans.md b/README-zh-Hans.md index 8edcff28bf04..b97b02f5ab84 100644 --- a/README-zh-Hans.md +++ b/README-zh-Hans.md @@ -5,10 +5,10 @@ Colossal-AI: 一个面向大模型时代的通用深度学习系统 -
- 用相同的硬件训练24倍大的模型
-- 超3倍的吞吐量
+- 超3倍的吞吐量
### BERT
@@ -145,7 +145,7 @@ Colossal-AI 为您提供了一系列并行组件。我们的目标是让您的
- [Open Pretrained Transformer (OPT)](https://github.com/facebookresearch/metaseq), 由Meta发布的1750亿语言模型,由于完全公开了预训练参数权重,因此促进了下游任务和应用部署的发展。
-- 加速45%,仅用几行代码以低成本微调OPT。[[样例]](https://github.com/hpcaitech/ColossalAI-Examples/tree/main/language/opt) [[在线推理]](https://service.colossalai.org/opt)
+- 加速45%,仅用几行代码以低成本微调OPT。[[样例]](https://github.com/hpcaitech/ColossalAI-Examples/tree/main/language/opt) [[在线推理]](https://service.colossalai.org/opt)
请访问我们的 [文档](https://www.colossalai.org/) 和 [例程](https://github.com/hpcaitech/ColossalAI-Examples) 以了解详情。
@@ -255,6 +255,28 @@ Colossal-AI 为您提供了一系列并行组件。我们的目标是让您的
## 安装
+### 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
+
+```bash
+pip install colossalai
+```
+
+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
+CUDA_EXT=1 pip install colossalai
+```
+
+**否则,PyTorch扩展只会在你实际需要使用他们时在运行时里被构建。**
+
+与此同时,我们也每周定时发布Nightly版本,这能让你提前体验到新的feature和bug fix。你可以通过以下命令安装Nightly版本。
+
+```bash
+pip install colossalai-nightly
+```
+
### 从官方安装
您可以访问我们[下载](https://www.colossalai.org/download)页面来安装Colossal-AI,在这个页面上发布的版本都预编译了CUDA扩展。
@@ -274,10 +296,10 @@ pip install -r requirements/requirements.txt
pip install .
```
-如果您不想安装和启用 CUDA 内核融合(使用融合优化器时强制安装):
+我们默认在`pip install`时不安装PyTorch扩展,而是在运行时临时编译,如果你想要提前安装这些扩展的话(在使用融合优化器时会用到),可以使用一下命令。
```shell
-NO_CUDA_EXT=1 pip install .
+CUDA_EXT=1 pip install .
```
(返回顶端)
diff --git a/README.md b/README.md index 1b0ca7e973e0..7aba907e0a64 100644 --- a/README.md +++ b/README.md @@ -257,9 +257,32 @@ Acceleration of [AlphaFold Protein Structure](https://alphafold.ebi.ac.uk/) ## Installation +### Install from PyPI + +You can easily install Colossal-AI with the following command. **By defualt, we do not build PyTorch extensions during installation.** + +```bash +pip install colossalai +``` + +However, if you want to build the PyTorch extensions during installation, you can set `CUDA_EXT=1`. + +```bash +CUDA_EXT=1 pip install colossalai +``` + +**Otherwise, CUDA kernels will be built during runtime when you actually need it.** + +We also keep release the nightly version to PyPI on a weekly basis. This allows you to access the unreleased features and bug fixes in the main branch. +Installation can be made via + +```bash +pip install colossalai-nightly +``` + ### Download From Official Releases -You can visit the [Download](https://www.colossalai.org/download) page to download Colossal-AI with pre-built CUDA extensions. +You can visit the [Download](https://www.colossalai.org/download) page to download Colossal-AI with pre-built PyTorch extensions. ### Download From Source @@ -270,9 +293,6 @@ You can visit the [Download](https://www.colossalai.org/download) page to downlo git clone https://github.com/hpcaitech/ColossalAI.git cd ColossalAI -# install dependency -pip install -r requirements/requirements.txt - # install colossalai pip install . ```