convirza/baseimages repository overview

Miniconda python 2.7 and 3.4 baseimages with ONBUILD instructions to copy app files to '/srv/app'. Tags for 2.7 and 3.4. These are maintained in the baseimages bitbucket⁠.

docker pull convirza/baseimages:py27
docker pull convirza/baseimages:py34
docker pull convirza/baseimages:py35

Or most likely if you're going to be actually using these:

FROM convirza/baseimages:py27
FROM convirza/baseimages:py34
FROM convirza/baseimages:py35

Becasue these are based on the miniconda python distribution there is no need for virtualenv. You should assume that the root environment is where all items should be installed. For scientific libraries, especially scikit learn, numpy, pandas, scipy, matplotlib, etc. use the conda distributions with:

conda install <package name>

Your requirements-conda.txt file should be conda formatted e.g.:

conda list -e

For everything else you can use the standard:

pip install <package name>

Your requirements.txt should be formatted for pip e.g.:

pip freeze