Window Docker TensorFlow Environment setup

Installing Docker

Before only the Docker file, not how to contact the installation of Docker environment, this time also try it, first download DockerToolbox.exe
After the installation is complete, the startup script start.sh, will default to check the version, if it is installed at the same time VirtualBox, it is recommended to restart, this card for a long time, has been reported to start vboxmanage abnormal, find a half day reason ...
This step on the Internet is still a lot of information, Docker related content is not familiar, dare not write the amount

To start a related image of TensorFlow

Execute the following command

Docker run-d-P 8888:8888--name= "tensor" gcr.io/tensorflow/tensorflow

-P 8,888:8,888 refers to the mapping of the local 8888 port to the 8888 port of the Docker image, the Jupyter service running in the tensor container occupies 8888 ports, this service is a Web-version editor that supports uploading, creating, modifying, and running Python programs.
--name is the name of the mirror
Gcr.io/tensorflow/tensorflow This is the official image of the TensorFlow, has integrated the TensorFlow and TensorFlow related dependencies

The first run will download the corresponding file from the server

Start the container

Docker exec-i tensor bash

You can also use the Docker-machine command to view the currently running Docker server

Docker-machine.exe ls

You can see that the IP of the Docker server is 192.1668.99.100

Access Http://192.168.99.100:8888/tree through the browser, is the Jupyter service, such as:

A password may be required for the first time, which is performed within the container's bash:

Jupyter Notebook List

Token is the password, the corresponding Python code can be uploaded through the page, and then perform the view within bash

Execute a simple program

Use Jupyter to execute, create a new program

Write the code and click Run to get the result

Appendix:

• Https://github.com/aymericdamien/TensorFlow-Examples?spm=5176.100239.blogcont60601.20.815eea70STfiF

Window Docker TensorFlow Environment setup