Regression: objectives and metrics


Name	Used for optimization	User-defined parameters	Formula and/or description
MAE	+	use_weights Default: true	Calculation principles
MAPE	+	use_weights Default: true	Calculation principles
Poisson	+	use_weights Default: true	Calculation principles
Quantile	+	use_weights Default: true alpha Default: 0.5	Calculation principles
RMSE	+	use_weights Default: true	Calculation principles
RMSEWithUncertainty	+	use_weights Default: true	Calculation principles
LogLinQuantile	+	use_weights Default: true alpha Default: 0.5	Calculation principles
Lq	+	use_weights Default: true q Default: Obligatory parameter	Calculation principles
Huber	+	delta Default: Obligatory parameter use_weights Default: true	Calculation principles
Expectile	+	use_weights Default: true alpha Default: 0.5	Calculation principles
Tweedie	+	use_weights Default: true variance_power Default: Obligatory parameter	Calculation principles
FairLoss	–	use_weights Default: true smoothness Default: 1.0	Calculation principles
NumErrors	–	greater_than use_weights Default: true	The proportion of predictions, for which the difference from the label value exceeds the specified value `greater_than`. Calculation principles
SMAPE	–	use_weights Default: true	Calculation principles
R2	–	use_weights Default: true	Calculation principles
MSLE	–	use_weights Default: true	Calculation principles
MedianAbsoluteError	–	–	Calculation principles


Name	Used for optimization	User-defined parameters	Formula and/or description
MAE	+	use_weights Default: true	Calculation principles
MAPE	+	use_weights Default: true	Calculation principles
Poisson	+	use_weights Default: true	Calculation principles
Quantile	+	use_weights Default: true alpha Default: 0.5	Calculation principles
RMSE	+	use_weights Default: true	Calculation principles
RMSEWithUncertainty	+	use_weights Default: true	Calculation principles
LogLinQuantile	+	use_weights Default: true alpha Default: 0.5	Calculation principles
Lq	+	use_weights Default: true q Default: Obligatory parameter	Calculation principles
Huber	+	delta Default: Obligatory parameter use_weights Default: true	Calculation principles
Expectile	+	use_weights Default: true alpha Default: 0.5	Calculation principles
Tweedie	+	use_weights Default: true variance_power Default: Obligatory parameter	Calculation principles
FairLoss	–	use_weights Default: true smoothness Default: 1.0	Calculation principles
NumErrors	–	greater_than use_weights Default: true	The proportion of predictions, for which the difference from the label value exceeds the specified value `greater_than`. Calculation principles
SMAPE	–	use_weights Default: true	Calculation principles
R2	–	use_weights Default: true	Calculation principles
MSLE	–	use_weights Default: true	Calculation principles
MedianAbsoluteError	–	–	Calculation principles