Public | Automated Build

Last pushed: a month ago
Short Description
Base image for data science projects
Full Description

docker-datascience

Data science-friendly Docker container

Goals:

  • Has all of the powerful interactive goodies that data scientists love.
  • Can run all production Even Python code.

Adding a new package

1) Adjust conda or pip commands in Dockerfile, or update requirements.txt
2) Run ./build_local.sh.
3) Rebuild downstream images (e.g., docker-compose build scholar-notebook.dev).
4) Once you're sure everything looks good, open a PR against this repo.
5) Don't commit anything to downstream repos until the image has been built and published.
6) Update the Even Base Environment in Domino.

Updating the Even Base environment in Domino

Do not forget to do this. Unfortunately, the Domino environment must be manually updated alongside this repo to match it. The two files are not identical; for example, Domino uses pip3 install <list of packages> instead of requirements.txt. Regardless, there are at least three things which must be religiously kept in sync:

  1. The protobuf version
  2. The exact python version, including the patch version.
  3. Exact python package versions, excepting local dev tools like Jupyter Notebook.
Docker Pull Command
Owner
evenco
Source Repository