logml.feature_importance.base

Classes

BaseImportanceExtractor([config, ...])

Base class for feature importance extraction.

class logml.feature_importance.base.BaseImportanceExtractor(config: Optional[object] = None, model_provider: Optional[logml.model_search.provider.ModelProvider] = None, n_models: Optional[int] = None, models: Optional[List[str]] = None, fallback_model: Optional[str] = None, global_params: Optional[dict] = None, logger=None, objective_config: Optional[logml.configuration.modeling.ModelingTaskSpec] = None)

Bases: object

Base class for feature importance extraction.

LABEL = None

CONFIG_CLASS = None

extract_model_feature_importance(model_name: Optional[str] = None, model_cls: Optional[Type[logml.models.base.BaseModel]] = None, params: Optional[dict] = None, dataset: Optional[logml.data.datasets.base.BaseDataset] = None, model: Optional[logml.models.base.BaseModel] = None): Feature importance extraction for single model.

extract_feature_importance(dataset: logml.data.datasets.base.BaseDataset, models: List[str]): Interface method for feature importance extraction.

get_raw_feature_importances() → dict: Raw importances.