forecaster.metrics.create_mean_pinball_loss

forecaster.metrics.create_mean_pinball_loss(alpha)

Create pinball loss for a given quantile.

Also known as quantile loss. Internally, it uses the mean_pinball_loss function from scikit-learn.

Parameters

Name	Type	Description	Default
alpha	float	Quantile for which the Pinball loss is calculated. Must be between 0 and 1, inclusive.	required

Returns

Name	Type	Description
	callable	Mean Pinball loss function for the given quantile.

Examples

import numpy as np
from spotforecast2_safe.forecaster.metrics import create_mean_pinball_loss

pinball_loss_50 = create_mean_pinball_loss(alpha=0.5)
y_true = np.array([1, 2, 3, 4, 5])
y_pred = np.array([1.1, 1.9, 3.2, 3.8, 5.1])
loss = pinball_loss_50(y_true, y_pred)
print(loss >= 0)

True