Copy the env.example to .env if you want postgres. The docker-compose just has some extras I found useful.

$ cp env.example .env
$ docker-compose up

or

$ docker build -t ts_stack
$ docker run ts_stack

This project is to do research on machine learning in a docker env. The toolkit to operate it is based off jupyter docker-stacks.

• https://github.com/jupyter/docker-stacks
• https://github.com/tensorflow

This thing works pretty well! Currently the python3 stuff is a little weird. If you get into terminal, you can probably make it work. I did once. But a lot of the TF examples don't do Py3 anyway.

I set up the java stuff to build out syntaxnet. Currently even just building from their sources fails with the same errrors. I'll update when I find the solution.

Cant distro Java cause Oracle. I think the breadcrumbs should be easy enough. If you have a solution on that and syntaxnet, let me know!

Also, You should really try pymux and pyvim in the Jupyter terminal. It's Glorious