shapiq.explainer

High-level interfaces for explaining ML model predictions.

The main entry point is Explainer, which auto-selects the best explanation method for a given model type. For more control, use one of the specialised explainers directly:

• TabularExplainer — tabular data via imputation-based games

• TabPFNExplainer — fast in-context learning with TabPFN

• AgnosticExplainer — any callable model

• TreeExplainer — optimised for tree-based models

See also

Examples & Tutorials for end-to-end examples.

Classes

Explainer	The main Explainer class for a simpler user interface.
TabularExplainer	The tabular explainer as the main interface for the shapiq package.
TabPFNExplainer	The TabPFN explainer as the main interface for the shapiq package.
AgnosticExplainer	Agnostic Explainer for shapiq.
TreeExplainer	The TreeExplainer class for tree-based models.