OCDocker.OCScore.Analysis.SHAP.Plots module

Utilities to visualize SHAP outputs.

Public helpers

• feature_importance_barh: horizontal bar chart of relative importance

• beeswarm: wrapper around shap.summary_plot

• shap_correlation_heatmap: correlation heatmap of SHAP values

Usage:

from OCDocker.OCScore.Analysis.SHAP.Plots import feature_importance_barh

OCDocker.OCScore.Analysis.SHAP.Plots.beeswarm(shap_2d, X_eval, out_png, figsize=(10, 6))

Wrapper around shap.summary_plot to save a beeswarm plot.

Parameters:

• shap_2d (np.ndarray) – SHAP values with shape (n_samples, n_features).

• X_eval (pd.DataFrame) – Evaluation features (columns = names).

• out_png (str) – Where to save the plot.

• figsize (tuple[int, int], optional) – Plot size (width, height). Default is (10, 6).

Returns:

Output path of the saved plot.

Return type:

str

OCDocker.OCScore.Analysis.SHAP.Plots.feature_importance_barh(shap_2d, feature_names, out_png, top_k=20, figsize=(10, 6))

Horizontal bar chart of relative SHAP importance per feature.

Parameters:

• shap_2d (np.ndarray) – SHAP values array with shape (n_samples, n_features).

• feature_names (list[str]) – Names of features in column order.

• out_png (str) – Where to save the plot.

• top_k (int, optional) – Number of top features to display. Default is 20.

• figsize (tuple[int, int], optional) – Figure size. Default is (10, 6).

Returns:

Output path of the saved plot.

Return type:

str

OCDocker.OCScore.Analysis.SHAP.Plots.shap_correlation_heatmap(shap_values, out_png, feature_names=None, figsize=(12, 10))

Plot a heatmap of SHAP value correlations across features.

Parameters:

• shap_values (array-like | pd.DataFrame) – SHAP values with shape (n_samples, n_features).

• out_png (str) – Where to save the heatmap image.

• feature_names (list[str] | None) – Optional feature names. If shap_values is a DataFrame, its columns are used.

• figsize (tuple[int, int]) – Figure size.

Returns:

Output path of the saved heatmap.

Return type:

str