zoom_in Click to enlarge

InterpretML

▲ ▼

language

description InterpretML Overview

InterpretML is a Python library focused on providing interpretable machine learning models. It allows users to build models that are inherently interpretable, rather than relying on post-hoc explanation techniques. InterpretML supports various model types, including generalized additive models (GAMs) and linear models, enabling users to understand the relationship between features and predictions.

help InterpretML FAQ

What is InterpretML used for?

InterpretML is an open-source Python toolkit for building and studying interpretable machine-learning models. It includes glassbox models that are designed to be understandable, rather than explaining a black-box model only after it has made a prediction.

What is an Explainable Boosting Machine in InterpretML?

An Explainable Boosting Machine, or EBM, is a generalized additive model that learns feature effects while keeping them viewable. InterpretML can show each feature's contribution, which makes an EBM useful for tabular risk or classification work.

Does InterpretML work with scikit-learn?

InterpretML models use familiar Python machine-learning workflows and can be compared with scikit-learn estimators. It also provides explainers for some black-box models, so it can fit into a scikit-learn-based experiment without replacing the whole pipeline.

Is InterpretML a replacement for SHAP?

No. InterpretML focuses strongly on interpretable models such as EBMs, while SHAP is mainly a framework for assigning feature contributions to predictions from many model types. They can be used together when a team wants both an interpretable baseline and post-hoc explanations.

swap_horiz

Looking for InterpretML alternatives? Compare top competitors ranked & scored

arrow_forward