interpret

Fit interpretable models. Explain blackbox machine learning.

What is interpret?

Fit interpretable models. Explain blackbox machine learning.

Overview of interpret

What it is

InterpretML is an open-source Python package that provides machine learning interpretability techniques. It allows users to train interpretable glassbox models and explain blackbox systems. The package is designed to help understand global model behavior or the reasons behind individual predictions.

Key features

  • Includes the Explainable Boosting Machine (EBM), an interpretable model developed at Microsoft Research that uses bagging, gradient boosting, and automatic interaction detection.
  • Supports glassbox models such as Explainable Boosting, APLR, Decision Tree, Decision Rule List, and Linear/Logistic Regression.
  • Supports blackbox explainers including SHAP Kernel Explainer, LIME, Morris Sensitivity Analysis, and Partial Dependence.
  • Provides Differentially Private EBMs for privacy-preserving model training.
  • Enables comparison of multiple model explanations.

Use cases

  • Model debugging to understand why a model made a mistake.
  • Feature engineering to improve model performance.
  • Detecting fairness issues in model predictions.
  • Human-AI cooperation to understand and trust model decisions.
  • Regulatory compliance for legal requirements.
  • High-risk applications in healthcare, finance, judicial, and other domains.

Getting started

shpip install interpret# ORconda install -c conda-forge interpret

License and pricing

The license is MIT. The package is open source and free to use.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/interpretmlinterpret.svg)](https://yunodb.com/tools/interpretmlinterpret?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/interpretmlinterpret?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/interpretmlinterpret.svg" alt="Featured on YunoDB" height="20"></a>