报错

ubuntu20.04安装NVIDIA Container Toolkit时执行sudo docker run --rm --gpus all nvidia/cuda:11.0.3-base-ubuntu20.04 nvidia-smi报错Unable to find image 'nvidia/cuda:11.0.3-base-ubuntu20.04' locally docker: Error response from daemon: Get "https://registry-1.docker.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers).

解决：

换源，参考教程：https://blog.csdn.net/qq_73162098/article/details/145014490?spm=1001.2014.3001.5506

但是这里面的源用不了，换成这里的源：

https://cloud.tencent.com/developer/article/2485043

docker安装

参考教程：https://blog.csdn.net/qq_73162098/article/details/145186905?spm=1001.2014.3001.5506

唯一一个跟着走没有报错的教程

但是我偷懒没换源，就导致安装NVIDIA Container Toolkit时执行到最后一步报错Unable to find image 'nvidia/cuda:11.0.3-base-ubuntu20.04' locally，于是又换源才解决报错

NVIDIA Container Toolkit安装

参考：https://blog.csdn.net/u011119817/article/details/110386179

报错Unable to find image 'nvidia/cuda:10.0-base' locally的话把最后一句改为：

sudo docker run --rm --gpus all nvidia/cuda:11.7.1-base-ubuntu20.04 nvidia-smi

或者是因为没有换源导致的

另外：报错permission一般是权限问题，在操作最前面加sudo就好