ComparisonReport.metrics#
- ComparisonReport.metrics()[source]#
Accessor for metrics-related operations.
You can access this accessor using the
metricsattribute.
Methods#
Here are the methods that are available through this accessor:
help()– Display accessor help using rich or HTMLsummarize()– Report a set of metrics for the estimatorsaccuracy()– Compute the accuracy scoreadd()– Add a custom metric to insummarize()available()– List available metric names in the registrybrier_score()– Compute the Brier scoreconfusion_matrix()– Plot the confusion matrixlog_loss()– Compute the log lossmae()– Compute the mean absolute errormape()– Compute the mean absolute percentage errorprecision()– Compute the precision scoreprecision_recall()– Plot the precision-recall curveprediction_error()– Plot the prediction error of a regression modelr2()– Compute the R² scorerecall()– Compute the recall scoreremove()– Remove a metric from each underlying estimator reportrmse()– Compute the root mean squared errorroc()– Plot the ROC curveroc_auc()– Compute the ROC AUC scoretimings()– Get all measured processing times related to the different estimators