InterpretML: A Unified Framework for Machine Learning Interpretability

TL;DR

InterpretML is an open-source Python package that unifies various machine learning interpretability methods, enabling easy comparison and visualization of glassbox models and blackbox explainability techniques, including the novel Explainable Boosting Machine.

Contribution

It introduces a unified API and visualization platform for interpretability algorithms and presents the first implementation of the Explainable Boosting Machine, a highly accurate interpretable model.

Findings

Provides a comprehensive, extensible interpretability toolkit

Includes the first implementation of the Explainable Boosting Machine

Enables easy comparison of interpretability methods

Abstract

InterpretML is an open-source Python package which exposes machine learning interpretability algorithms to practitioners and researchers. InterpretML exposes two types of interpretability - glassbox models, which are machine learning models designed for interpretability (ex: linear models, rule lists, generalized additive models), and blackbox explainability techniques for explaining existing systems (ex: Partial Dependence, LIME). The package enables practitioners to easily compare interpretability algorithms by exposing multiple methods under a unified API, and by having a built-in, extensible visualization platform. InterpretML also includes the first implementation of the Explainable Boosting Machine, a powerful, interpretable, glassbox model that can be as accurate as many blackbox models. The MIT licensed source code can be downloaded from github.com/microsoft/interpret.