docker使用，包括有nvidia-docker

docker run 命令：https://www.runoob.com/docker/docker-run-command.html

docker run -it ubuntu:15.10 /bin/bash #指定ubuntu15.04为镜像
docker run -it -v /home/work/suwenyuan:/home/work/suwenyuan  --name antonio_ubuntu16. ubuntu16.04_dl_platform /bin/bash 
nvidia-docker run -it -v /home/work/chenpeili/:/home/work/chenpeili/ --runtime=nvidia -e NVIDIA_VISIBLE_DEVICES=all --name anotonio_last_one gw000/keras:2.1.4-py3-tf-gpu /bin/bash

docker images

docker  ps -a#查看容器

重新命名镜像名

docker tag 657d80a6401d  ubuntu16.04_dl_platform# 657d80a6401d  是镜像id

Error response from daemon: Container ************** is not running

docker start 1cd6f9a81186
docker exec -it 1cd6f9a81186 /bin/bash #重新进入容器，1cd6f9a81186是容器id

可以使用的镜像

tensorflow/tensorflow:1.12.-devel-gpu-py3

退出容器保持后台运行 Ctrl + P + Q