ISLR-cauldron

An Introduction to Statistical Learning (James, Witten, Hastie, Tibshirani, 2013): Python code. Inspired by https://github.com/JWarmenhoven/ISLR-python

This repository has all the exercises the An Introduction to Statistical Learning has in R but in python. The difference of this repository from ISLR-python is that instead of using Jupiter Notebooks, it will use Cauldron, the unnotebook.

Cauldron provides a lot of benefits when compared to Jupiter Notebooks, it allows the user to:

• Use his preferred editor. This will allow the user to get all the benefits from editors like Pycharm.
• Code reviews are made easy. Instead of using blob file, it uses .py files. This will allow a smooth code review.
• Cauldron is also ready for production. The user can use it's CLI to deploy their notebooks to production.

Check all the benefits on Cauldron site.

Cauldron is free and open source!

Run Notebooks in Docker Container

You can use the docker image in this repo to run your notebooks. This way you avoid having to install all the python libraries locally.

This is how you start your container and run your notebooks:

• Run docker-compose up to start container: docker-compose up
• Start Cauldron with http://127.0.0.1:5010

Installing new python libraries

• Modify the requirements.txt with the additional libraries.
• Rebuild the container

docker-compose build