This repository will house explorations in machine_learning.
The Dockerfile contains a docker image that can be used to run these notebooks. The docker will install scikit-learn, xgboost, numpy, scipy, theano, tensorflow, keras, and the jupyter notebook. If you don't have Docker, you can get installation instructions here.
Clone this repository onto your local machine. We'll call this directory LOCAL_DIR, which on my machine is /Users/dxwils3/git/machine_learning.
To install the Docker image, run
docker run -d -p 8888:8888 -v LOCAL_DIR:/notebook dxwils3/machine_learning, which will pull the pre-built image from docker hub.
This will mount the directory LOCAL_DIR as /notebook in the container, which will be the jupyter notebook home directory.
To get to the jupyter notebook, visit localhost:8888 on Unix or Windows, or dockerhost:8888 on Mac OSX, assuming you are using docker-osx-dev.
To get a shell on this running docker, use docker exec -i -t CONTAINER_ID bash where container id is given by executing docker ps.