问题复现
docker 执行docker file文件时报错：

ERROR: failed to do request: Head https://registry-1.docker.io/v2/pytorch/pytorch/manifests/1.6.0-cuda10.1-cudnn7-devel: dial tcp: lookup registry-1.docker.io on 192.168.65.5:53: no such host

[internal] load metadata for docker.io/pytorch/pytorch:1.6.0-cuda10.1-cudnn7-devel:

failed to solve with frontend dockerfile.v0: failed to create LLB definition: failed to do request: Head https://registry-1.docker.io/v2/pytorch/pytorch/manifests/1.6.0-cuda10.1-cudnn7-devel: dial tcp: lookup registry-1.docker.io on 192.168.65.5:53: no such host

或者，docker拉取镜像时报错：

Error response from daemon: Head https://registry-1.docker.io/v2/nvidia/cuda/manifests/10.1-cudnn7-devel-ubuntu16.04: Get https://auth.docker.io/token?account=yang929604665&scope=repository%3Anvidia%2Fcuda%3Apull&service=registry.docker.io: EOF

原因：
鉴于国内网络问题，后续拉取 Docker 镜像十分缓慢，需要配置加速器来解决，可以使用的是网易的镜像地址：http://hub-mirror.c.163.com。

解决办法：
在

C:\Users\用户名\.docker

路径的
daemon.json文件中，修改为

{
  "registry-mirrors": [
    "http://hub-mirror.c.163.com",
    "https://docker.mirrors.ustc.edu.cn/"
  ],
  "insecure-registries": [],
  "debug": false,
  "experimental": false,
  "features": {
    "buildkit": true
  }
}

最后重启docker即可。