saealib.LOOAccuracyEvaluator#

class saealib.LOOAccuracyEvaluator(metrics=None)[ソース]#

ベースクラス: AccuracyEvaluator

Evaluates surrogate accuracy via leave-one-out cross-validation.

Equivalent to KFoldAccuracyEvaluator with n_splits=n_samples. Each sample is used as a validation set exactly once.

パラメータ:

metrics (list[SurrogateAccuracyMetric] or None) -- Metrics to compute.

Methods

__init__

evaluate

Evaluate via leave-one-out CV.

Method Details

LOOAccuracyEvaluator.__init__(metrics=None)#
パラメータ:

metrics (list[SurrogateAccuracyMetric] | None)

LOOAccuracyEvaluator.evaluate(surrogate, train_x, train_y)[ソース]#

Evaluate via leave-one-out CV.

パラメータ:

  • surrogate (Surrogate)

  • train_x (np.ndarray)

  • train_y (np.ndarray)

戻り値の型:

SurrogateAccuracy