How do I run a python script with Tensorflow running in a Docker on Windows?

Question

Imagine I manage to install Tensorflow on Windows, using Docker as in these two links, for example:

TensorFlow on Windows

How to install and run TensorFlow on a Windows PC

In both links, they're able to use Tensorflow on a shell python (don't know exactly what version, I have Anaconda installed).

But what if I want to run a script that I made on my local machine that has Tensorflow in it? How do I call the script from docker? I mean, how do I find the script (located on my desktop, for instance) from docker to run it?

Answer 1

If you want your container (that has Tensorflow already preinstalled, since it is running from the Tensorflow image) to access your script, you need to mount that script from your host onto a local path in your container.

docker run -v /path/to/your/script:/path/to/script

See "Mount a host file as a data volume".

The -v flag can also be used to mount a single file - instead of just directories - from the host machine.

$ docker run --rm -it -v ~/.bash_history:/.bash_history ubuntu /bin/bash

Then, from your container, you will access the same script in /path/to/script.

Alex Pryiomka gives an example of running such a script in tensorflow with "How to run Python Scripts on Mac Terminal using Docker with Tensorflow?"