在WSL中配置GPU环境

﹏ヽ暗。殇╰゛Y 2023-09-25 14:08 116阅读 0赞

首先需要明确一点，虽然我们通过安装WSL获得了linux开发环境，但是我们最终使用的GPU还是在windows当中的，所以还是需要在系统中安装对应的驱动。

#### 第一步：在window上根据显卡型号和版本安装驱动 ####

这里参考之前的步骤就行

#### 第二步：安装WSL ####

参考官方文档即可

[安装 WSL![icon-default.png_t_M85B][]https://link.zhihu.com/?target=https%3A//learn.microsoft.com/zh-cn/windows/wsl/install][WSL_icon-default.png_t_M85B_https_link.zhihu.com_target_https_3A_learn.microsoft.com_zh-cn_windows_wsl_install]

#### 第三步：在WSL中安装CUDA Toolkit ####

以下是针对Ubuntu18.04的url

sudo apt-key adv --fetch-keys http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub
    sudo sh -c 'echo "deb Index of /compute/cuda/repos/ubuntu1804/x86_64 /" > /etc/apt/sources.list.d/cuda.list'
    sudo apt-get update
    sudo apt-get install -y cuda-toolkit-11-0

其他版本的可以在官网上找：[CUDA Toolkit][]

CUDAToolkit版本和驱动版本的关系：[https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html][https_docs.nvidia.com_cuda_cuda-toolkit-release-notes_index.html]

![format_png][]

wget https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/cuda-wsl-ubuntu.pin
    sudo mv cuda-wsl-ubuntu.pin /etc/apt/preferences.d/cuda-repository-pin-600
    wget https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda-repo-wsl-ubuntu-12-0-local_12.0.0-1_amd64.deb
    sudo dpkg -i cuda-repo-wsl-ubuntu-12-0-local_12.0.0-1_amd64.deb
    sudo cp /var/cuda-repo-wsl-ubuntu-12-0-local/cuda-*-keyring.gpg /usr/share/keyrings/
    sudo apt-get update
    sudo apt-get -y install cuda

安装 nvidia-cuda-toolkit

sudo apt install nvidia-cuda-toolkit

检查是否安装成功：

nvcc -V
    # 能看到对应版本就行

#### 第四步：安装pytorch ####

选择对应版本的pytorch即可：

[PyTorchpytorch.org/get-started/locally/pytorch.org/get-started/locally/][PyTorchpytorch.org_get-started_locally_pytorch.org_get-started_locally]

> 在这里有一个大坑，安装pytorch的时候不能选择conda的安装链接，会自动给你安装默认版本的，然后导致版本不对应，所以一定要采用pip的方式来安装，一定要采用pip的方式来安装，一定要采用pip的方式来安装

[在WSL中配置GPU环境 - 知乎][WSL_GPU_ -]

[icon-default.png_t_M85B]: https://csdnimg.cn/release/blog_editor_html/release2.2.0/ckeditor/plugins/CsdnLink/icons/icon-default.png?t=M85B
[WSL_icon-default.png_t_M85B_https_link.zhihu.com_target_https_3A_learn.microsoft.com_zh-cn_windows_wsl_install]: https://link.zhihu.com/?target=https%3A//learn.microsoft.com/zh-cn/windows/wsl/install
[CUDA Toolkit]: https://link.zhihu.com/?target=https%3A//developer.nvidia.com/cuda-downloads
[https_docs.nvidia.com_cuda_cuda-toolkit-release-notes_index.html]: https://link.zhihu.com/?target=https%3A//docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html
[format_png]: https://img-blog.csdnimg.cn/img_convert/c5279a76fcc1d8d4bd10fffd5b85a3c3.webp?x-oss-process=image/format,png
[PyTorchpytorch.org_get-started_locally_pytorch.org_get-started_locally]: https://link.zhihu.com/?target=https%3A//pytorch.org/get-started/locally/
[WSL_GPU_ -]: https://zhuanlan.zhihu.com/p/565653129