ML.EVAL.REGRESSION.D2_PINBALL_SCORE

Returns the D^2 regression score function, fraction of pinball loss explained.

Syntax

ML.EVAL.REGRESSION.D2_PINBALL_SCORE(y_true, y_pred, sample_weight, alpha, multioutput)

Arguments

Name	Type	Default	Description
y_true	object		DataFrame or array object of ground-truth target values.
y_pred	object		DataFrame or array object of predicted target values.
sample_weight	Any	None	Optional DataFrame or array object of per-sample weights. Omit for uniform weights.
alpha	Any	0.5	Slope of the pinball loss between 0 and 1. 0.5 is equivalent to the mean absolute error.
multioutput	Any	"uniform_average"	How to aggregate scores across multiple output dimensions. One of 'raw_values', 'uniform_average', or 'variance_weighted'.

Examples

Examples coming soon

Working Excel formula examples for this function are not yet written.

See also

• ML.EVAL.REGRESSION.D2_ABSOLUTE_ERROR_SCORE
• ML.EVAL.REGRESSION.D2_TWEEDIE_SCORE
• ML.EVAL.REGRESSION.EXPLAINED_VARIANCE_SCORE
• ML.EVAL.REGRESSION.MAX_ERROR
• ML.EVAL.REGRESSION.MEAN_ABSOLUTE_ERROR
• ML.EVAL.REGRESSION.MEAN_ABSOLUTE_PERCENTAGE_ERROR
• ML.EVAL.REGRESSION.MEAN_GAMMA_DEVIANCE
• ML.EVAL.REGRESSION.MEAN_POISSON_DEVIANCE
• ML.EVAL.REGRESSION.MEAN_SQUARED_ERROR
• ML.EVAL.REGRESSION.MEAN_SQUARED_LOG_ERROR
• ML.EVAL.REGRESSION.MEDIAN_ABSOLUTE_ERROR
• ML.EVAL.REGRESSION.R2_SCORE
• ML.EVAL.REGRESSION.ROOT_MEAN_SQUARED_ERROR
• ML.EVAL.REGRESSION.ROOT_MEAN_SQUARED_LOG_ERROR