pysad.core.BaseMetric

class pysad.core.BaseMetric

Abstract base class for metrics.

Methods

__init__()
get()	Gets the current value of the score.
update(y_true, y_pred)	Updates the metric with given true and predicted value for a timestep.

abstract get()

Gets the current value of the score. Note that some methods such as AUPR and AUROC gives exception when used with only one class exist in the list of previous y_trues.

Returns:

The current score.

Return type:

float

abstract update(y_true, y_pred)

Updates the metric with given true and predicted value for a timestep.

Parameters:

• y_true (int) – Ground truth class. Either 1 or 0.

• y_pred (float) – Predicted class or anomaly score. Higher values correspond to more anomalousness and lower values correspond to more normalness.