doker 使用英伟达GPU报错处理

近在docker容器中部署使用英伟达GPU的应用出现以下报错：

Error response from daemon: could not select device driver "nvidia" with capabilities: [[gpu]]

原因为：

nvidia-container-toolkit 未安装

ubuntu环境下的解决方法如下：

1、更新apt 仓库

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
  && curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
    sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
    sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

2、更新仓库

 sudo apt-get update

3、安装

sudo apt-get install -y nvidia-container-toolkit

4、验证

which nvidia-container-runtime

输出 /usr/bin/nvidia-container-runtime，表示安装成功。

报错消失

其他环境参考