Averaging Methods¶ MacroAverage ¶ Bases: Averaging Computes the arithmetic mean over all classes, also known as macro-averaging. aliases = ('macro', 'macro_average', 'mean') ¶ dependencies = () ¶ sklearn_equivalent = 'macro' ¶ WeightedAverage ¶ Bases: Averaging Computes the prevalence weighted mean over all classes, also known as weighted averaging. aliases = ('weighted', 'weighted_average', 'micro', 'micro_average') ¶ dependencies = ('p_condition',) ¶ sklearn_equivalent = 'weighted' ¶ SelectPositiveClass ¶ Bases: Averaging Selects only the positive class, also known as binary averaging. Parameters: positive_class (int, default: -1 ) – the class to use as the average. Defaults to -1, the last class. aliases = ('select_positive', 'binary', 'select') ¶ dependencies = () ¶ sklearn_equivalent = 'binary' ¶ HarmonicMean ¶ Bases: Averaging Computes the harmonic mean over all classes. aliases = ('harmonic', 'harm') ¶ dependencies = () ¶ sklearn_equivalent = None ¶ GeometricMean ¶ Bases: Averaging Computes the geometric mean over all classes. aliases = ('geometric', 'geom') ¶ dependencies = () ¶ sklearn_equivalent = None ¶ Minimum ¶ Bases: Averaging Computes the minimum over all classes. aliases = ('min', 'minimum') ¶ dependencies = () ¶ sklearn_equivalent = None ¶ Maximum ¶ Bases: Averaging Computes the maximum over all classes. aliases = ('max', 'maximum') ¶ dependencies = () ¶ sklearn_equivalent = None ¶