Interpretability implementations + demos

Code for implementations of interpretable machine learning models and demos of how to use various interpretability techniques (with accompanying slides here).

Code implementations

Provides scikit-learn style wrappers/implementations of different interpretable models (see readmes in individual folders within imodels for details)

• bayesian rule lists
• optimal classification tree
• sparse integer linear models (simple, unstable implementation)

Demo notebooks

The demos are contained in 3 main notebooks, summarized in cheat_sheet.pdf

1. model_based.ipynb - how to use different interpretable models
2. posthoc.ipynb - different simple analyses to interpret a trained model
3. uncertainty.ipynb - code to get uncertainty estimates for a model

Installation / quickstart

The interpretable models within the imodels folder can be easily installed and used.

pip install git+https://github.com/csinva/interpretability-implementations-demos

from imodels.bayesian_rule_lists import RuleListClassifier
model = RuleListClassifier()
model.fit(X_train, y_train)
model.score(X_test, y_test)
preds = model.predict(X_test)

References / further reading

• book on interpretable machine learning
• high-level review on interpretable machine learning
• review on black-blox explanation methods
• review on variable importance