interpret_community.adapter.explanation_adapter module

Defines an adapter for creating an interpret-community style explanation from other frameworks.

class interpret_community.adapter.explanation_adapter.ExplanationAdapter(features=None, classification=False, method='Adapter')

Bases: object

An adapter for creating an interpret-community explanation from local importance values.

Parameters

  • features (list[str]) – A list of feature names.

  • classification (bool) – Indicates if this is a classification or regression explanation.

  • method (str) – The explanation method used to explain the model (e.g., SHAP, LIME).

create_global(local_importance_values, evaluation_examples=None, expected_values=None, include_local=True, batch_size=100)

Create a global explanation from the list of local feature importance values.

Parameters

  • local_importance_values (numpy.ndarray or scipy.sparse.csr_matrix or list[scipy.sparse.csr_matrix]) – The feature importance values.

  • evaluation_examples (numpy.ndarray or pandas.DataFrame or scipy.sparse.csr_matrix) – A matrix of feature vector examples (# examples x # features) on which to explain the model’s output.

  • expected_values (numpy.ndarray) – The expected values of the model.

  • include_local (bool) – Include the local explanations in the returned global explanation. If include_local is False, will stream the local explanations to aggregate to global.

  • batch_size (int) – If include_local is False, specifies the batch size for aggregating local explanations to global.

create_local(local_importance_values, evaluation_examples=None, expected_values=None)

Create a local explanation from the list of local feature importance values.

Parameters