In this repository, we explore how to understand decision trees and random forests better. This serves to provide additional interpretation beyond just the feature importances. We do this by using the treeinterpreter package at https://github.com/andosa/treeinterpreter. This package looks at the individual contribution of each feature for a given example.
Links to his blog:
- http://blog.datadive.net/interpreting-random-forests/
- http://blog.datadive.net/random-forest-interpretation-with-scikit-learn/
- http://blog.datadive.net/random-forest-interpretation-conditional-feature-contributions/
View my blog on this topic here.