PDPProb

class ordinal_xai.interpretation.pdp_prob.PDPProb(model: Any, X: DataFrame, y: Series | None = None)[source]

Bases: BaseInterpretation

Partial Dependence Plot (PDP) interpretation method for ordinal regression models. This version visualizes class probabilities as stacked area plots.

__init__(model: Any, X: DataFrame, y: Series | None = None)[source]

Initialize the PDPProb interpretation method.

Parameters:

  • model (BaseOrdinalModel) – The trained ordinal regression model.

  • X (pd.DataFrame) – DataFrame containing the dataset used for interpretation.

  • y (Optional[pd.Series], default=None) – Series containing target labels.

explain(observation_idx: int | None = None, feature_subset: List[str] | List[int] | None = None, plot: bool = False, max_features_per_figure: int = 12) → Dict[str, Dict[str, ndarray]][source]

Generate Partial Dependence Plots as stacked area plots for class probabilities.

Parameters:

  • observation_idx (Optional[int], default=None) – Ignored (PDP is a global method).

  • feature_subset (Optional[Union[List[str], List[int]]], default=None) – List of feature names or indices to plot. If None, all features are used.

  • plot (bool, default=False) – Whether to create visualizations.

  • max_features_per_figure (int, default=12) – Maximum number of features to display per figure (for large datasets).

Returns:

Dictionary containing PDP results for each feature: - ‘grid_values’: Feature values used for plotting - ‘average’: Average predicted probabilities at each feature value

Return type:

Dict[str, Dict[str, np.ndarray]]

_plot_pdps_prob(results: Dict[str, Dict[str, ndarray]], feature_subset: List[str], max_features_per_figure: int = 12) → None[source]

Create visualization of PDP probability plots with pagination for large feature sets.

Parameters:

  • results (Dict[str, Dict[str, np.ndarray]]) – PDP results for each feature

  • feature_subset (List[str]) – List of features to plot

  • max_features_per_figure (int, default=12) – Maximum number of features to display per figure

_abc_impl = <_abc._abc_data object>