Get logic explanations from the neural model

torch_explain.logic.nn

torch_explain.logic.nn.explain_class()

Generate a local explanation for a single sample.

Parameters

• model – pytorch model

• x – input samples to extract logic formulas.

• y1h – target labels to extract logic formulas (MUST be one-hot encoded).

• x_val – input samples to validate logic formulas.

• y_val1h – target labels to validate logic formulas (MUST be one-hot encoded).

• target_class – target class.

• max_minterm_complexity – maximum number of concepts per logic formula (per sample).

• topk_explanations – number of local explanations to be combined.

• max_accuracy – if True a formula is simplified only if the simplified formula gets 100% accuracy.

• concept_names – list containing the names of the input concepts.

Returns

Local explanation