This tutorial requires Python 3.5+ as well as,

• scikit-learn >=0.21.2
• matplotlib
• pandas

The tutorial notebook can be found in new-features-tutorial/new-features-tutorial.ipynb.

This interpretability tutorial uses some features from scikit-learn which are under-development:

• DataFrame handling with OpenML datasets: scikit-learn/scikit-learn#13902
• Fast partial dependence plot for Gradient Boosting Decsision Trees: scikit-learn/scikit-learn#13769
• Permutation feature importance: scikit-learn/scikit-learn#13146

These features have been combined into a scikit-learn branch in the following repository: https://github.com/glemaitre/scikit-learn/tree/workshop

You can refer to the following documentation to install scikit-learn from such source: https://scikit-learn.org/stable/developers/advanced_installation.html#install-bleeding-edge