plot_shap_decision

Note

This page is a reference documentation. It only explains the function signature, and not how to use it. Please refer to the Habitat Guide and Python API guide (v2.0) for usage.

plot_shap_decision(shap_values: ndarray, features: ndarray, *, feature_names: Sequence[str] | None = None, sample_indices: Sequence[int] | None = None, base_value: float = 0.0, title: str = 'SHAP decision')[source]

SHAP decision paths for a handful of samples.

Parameters:

  • shap_values – Attribution matrix (n_samples, n_features).

  • features – Feature matrix aligned with shap_values.

  • feature_names – Optional column names.

  • sample_indices – Rows to draw. Default: up to 10 evenly spaced samples across the SHAP-sum range.

  • base_value – Explainer base value (positive class).

  • title – Figure title (sanitised).

Returns:

The matplotlib Figure.