logml.metrics.base

Classes

BaseMetric(**kwargs)

Base class for LogML metrics.

class logml.metrics.base.BaseMetric(**kwargs)

Bases: object

Base class for LogML metrics.

LABEL: str = None

TASK: logml.common.ModelingTask = None

REQUIRED: bool = True

is_applicable(objective: logml.common.ModelingTask, y_true: Optional[numpy.ndarray] = None, y_pred: Optional[numpy.ndarray] = None, y_pred_proba: Optional[numpy.ndarray] = None, return_loss: Optional[bool] = None, return_score: Optional[bool] = None, **kwargs) → bool: Checks if the metric is applible for the modeling problem at hand.

get_value(y_true: Optional[numpy.ndarray] = None, y_pred: Optional[numpy.ndarray] = None, y_pred_proba: Optional[numpy.ndarray] = None, class_labels: Optional[list] = None, **kwargs): Get metric value

get_score(**kwargs): Get metric value as score (the larger the better).

get_loss(**kwargs): Get metric value as loss (the smaller the better).

get_scoring_fn() → collections.abc.Callable: Returns standalone scoring function to be used in estimators, like PermutationFeatureExtractor.