calculate_feature_importance#
- calculate_feature_importance(model: Any, dataset: Union[Dataset, DataFrame], force_permutation: bool = False, permutation_kwargs: Optional[Dict[str, Any]] = None) Tuple[Series, str][source]#
Calculate features effect on the label.
- Parameters
- modelt.Any
a fitted model
- datasett.Union[‘tabular.Dataset’, pd.DataFrame]
dataset used to fit the model
- force_permutationbool, default: False
force permutation importance calculation
- permutation_kwargst.Dict[str, t.Any] , default: None
kwargs for permutation importance calculation
- Returns
- Tuple[Series, str]:
first item - feature importance normalized to 0-1 indexed by feature names, second item - type of feature importance calculation (types: permutation_importance, feature_importances_, coef_)
- Raises
- NotFittedError
Call ‘fit’ with appropriate arguments before using this estimator.
- DeepchecksValueError
if model validation failed. if it was not possible to calculate features importance.
- NumberOfFeaturesLimitError
if the number of features limit were exceeded.