SHAP

SHAP (SHapley Additive exPlanations) is a game theoretic approach to explain the output of any machine learning model. It connects optimal credit allocation with local explanations using the classic Shapley values from game theory and their related extensions.

More Information

For more information visit the SHAP website.

Available Versions

These versions of SHAP are available on the BEAR systems (BlueBEAR and BEAR Cloud VMs). These will be retained in accordance with our Applications Support and Retention Policy.