classification

class aitoolbox.experiment.core_metrics.classification.AccuracyMetric(y_true, y_predicted, positive_class_thresh=0.5)

Bases: AbstractBaseMetric

Model prediction accuracy

Parameters:

• y_true (numpy.ndarray or list) – ground truth targets

• y_predicted (numpy.ndarray or list) – predicted targets

• positive_class_thresh (float or None) – predicted probability positive class threshold. Set it to None when dealing with multi-class labels.

calculate_metric()

Perform metric calculation and return it from this function

Returns:

return metric_result

Return type:

float or dict

class aitoolbox.experiment.core_metrics.classification.ROCAUCMetric(y_true, y_predicted)

Bases: AbstractBaseMetric

Model prediction ROC-AUC

Parameters:

• y_true (numpy.ndarray or list) – ground truth targets

• y_predicted (numpy.ndarray or list) – predicted targets

calculate_metric()

Perform metric calculation and return it from this function

Returns:

return metric_result

Return type:

float or dict

class aitoolbox.experiment.core_metrics.classification.PrecisionRecallCurveAUCMetric(y_true, y_predicted)

Bases: AbstractBaseMetric

Model prediction PR-AUC

Parameters:

• y_true (numpy.ndarray or list) – ground truth targets

• y_predicted (numpy.ndarray or list) – predicted targets

calculate_metric()

Perform metric calculation and return it from this function

Returns:

return metric_result

Return type:

float or dict

class aitoolbox.experiment.core_metrics.classification.F1ScoreMetric(y_true, y_predicted, positive_class_thresh=0.5)

Bases: AbstractBaseMetric

Model prediction F1 score

Parameters:

• y_true (numpy.ndarray or list) – ground truth targets

• y_predicted (numpy.ndarray or list) – predicted targets

• positive_class_thresh (float) – predicted probability positive class threshold

calculate_metric()

Perform metric calculation and return it from this function

Returns:

return metric_result

Return type:

float or dict

class aitoolbox.experiment.core_metrics.classification.PrecisionMetric(y_true, y_predicted, positive_class_thresh=0.5)

Bases: AbstractBaseMetric

Model prediction precision

Parameters:

• y_true (numpy.ndarray or list) – ground truth targets

• y_predicted (numpy.ndarray or list) – predicted targets

• positive_class_thresh (float) – predicted probability positive class threshold

calculate_metric()

Perform metric calculation and return it from this function

Returns:

return metric_result

Return type:

float or dict

class aitoolbox.experiment.core_metrics.classification.RecallMetric(y_true, y_predicted, positive_class_thresh=0.5)

Bases: AbstractBaseMetric

Model prediction recall score

Parameters:

• y_true (numpy.ndarray or list) – ground truth targets

• y_predicted (numpy.ndarray or list) – predicted targets

• positive_class_thresh (float) – predicted probability positive class threshold

calculate_metric()

Perform metric calculation and return it from this function

Returns:

return metric_result

Return type:

float or dict