Python docker for datascience
An extension to the Kaggle python docker for data science. The prebuilt version of this extension is at sharif1093/python-datascience-docker. Useful links:
- Place of
- Place of the
Docker Hub: kaggle/python
- Tutorial on using the main
To install the prebuilt docker image from http://www.docker.com, use the following instaructions:
docker run -d --name python-server -v ADDRESS/TO/YOUR/PYTHON/WORKSPACE:/tmp/working -p 8888:8888 --restart=always sharif1093/python-datascience-docker
Then to access the Jupiter notebook on the docker, type the following in your browser:
You will have access to it, locally and remotely.
To see the activity logs of the running server, run:
docker logs -f python-server
To access to the server shell, run:
docker exec -it python-server bash
To stop the server run:
docker stop python-server
And to start it again run:
docker start python-server
And finally, to remove the docker image while it is stopped, run:
docker rm python-server docker rmi sharif1093/python-datascience-docker
Setting up custom Jupyter server
Read this document to learn more about setting up a public Jupyter server.
Guide on using Jupyter
There are many nice references on using Jupyter. You may find one of them here. An amazing set of Jupyter notebooks might be found here. This is the pdf version of the Jupyter doc on readthedocs. This guide on tips & tricks of Jupyter is also interesting to know.