Class AggregateClassificationMetrics (1.25.0)

Aggregate metrics for classification/classifier models. For multi-class models, the metrics are either macro-averaged or micro-averaged. When macro-averaged, the metrics are calculated for each label and then an unweighted average is taken of those values. When micro-averaged, the metric is calculated globally by counting the total number of correctly predicted rows.

Recall is the fraction of actual positive labels that were given a positive prediction. For multiclass this is a macro- averaged metric.

Threshold at which the metrics are computed. For binary classification models this is the positive class threshold. For multi-class classfication models this is the confidence threshold.

Logarithmic Loss. For multiclass this is a macro-averaged metric.

Inheritance

builtins.object > google.protobuf.pyext._message.CMessage > builtins.object > google.protobuf.message.Message > AggregateClassificationMetrics