make_forecasting_scorer#

make_forecasting_scorer(func, name=None, greater_is_better=False, multioutput='uniform_average', multilevel='uniform_average')[source]#

Create a metric class from a metric functions.

Parameters
func

Function to convert to a forecasting scorer class. Score function (or loss function) with signature func(y, y_pred, **kwargs).

namestr, default=None

Name to use for the forecasting scorer loss class.

greater_is_betterbool, default=False

If True then maximizing the metric is better. If False then minimizing the metric is better.

multioutput{‘raw_values’, ‘uniform_average’} or array-like of shape (n_outputs,), default=’uniform_average’

Defines how to aggregate metric for multivariate (multioutput) data. If array-like, values used as weights to average the errors. If ‘raw_values’, returns a full set of errors in case of multioutput input. If ‘uniform_average’, errors of all outputs are averaged with uniform weight.

multilevel{‘raw_values’, ‘uniform_average’}

Defines how to aggregate metric for hierarchical data (with levels). If ‘uniform_average’ (default), errors are mean-averaged across levels. If ‘raw_values’, does not average errors across levels, hierarchy is retained.

Returns
scorer:

Metric class that can be used as forecasting scorer.