Tensorflow环境安装和使用（通过docker）

tensorflow有一种简单的使用方法，就是直接运行官方提供的docker镜像，该镜像安装好了tensorflow运行环境，直接食用即可：

sudo docker pull tensorflow/tensorflow:latest

sudo docker run -it -p 8888:8888 tensorflow/tensorflow:latest-jupyter

输出如下：

[xxx@fedora ~]$ docker run -it -p 8888:8888 tensorflow/tensorflow:latest-jupyter

[I 02:53:45.331 NotebookApp] Writing notebook server cookie secret to /root/.local/share/jupyter/runtime/notebook_cookie_secret

jupyter_http_over_ws extension initialized. Listening on /http_over_websocket

[I 02:53:45.591 NotebookApp] Serving notebooks from local directory: /tf

[I 02:53:45.591 NotebookApp] Jupyter Notebook 6.3.0 is running at:

[I 02:53:45.591 NotebookApp] http://5da7cc9a6978:8888/?token=4865b5c6fbe5ffcf5ea6a2c645f0a6dd246d55ac1f57b0fd

[I 02:53:45.591 NotebookApp] or http://127.0.0.1:8888/?token=4865b5c6fbe5ffcf5ea6a2c645f0a6dd246d55ac1f57b0fd

[I 02:53:45.591 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).

[C 02:53:45.598 NotebookApp]

To access the notebook, open this file in a browser:

file:///root/.local/share/jupyter/runtime/nbserver-1-open.html

Or copy and paste one of these URLs:

http://5da7cc9a6978:8888/?token=4865b5c6fbe5ffcf5ea6a2c645f0a6dd246d55ac1f57b0fd

or http://127.0.0.1:8888/?token=4865b5c6fbe5ffcf5ea6a2c645f0a6dd246d55ac1f57b0fd

上一步骤开启jupyter服务时，最后提供了登录地址，通过浏览器访问即可。

5、使用指定版本的tensorflow

latest是最新的版本，如果需要使用特定版本的tensorflow，就需要对应版本的镜像。tensorflow的镜像仓库地址如下，可以搜索查找对应的tag版本：

https://hub.docker.com/r/tensorflow/tensorflow/tags?page=1&ordering=last_updated

(End)