holisticai.explainability.metrics.surrogate_mean_squared_error_degradation

holisticai.explainability.metrics.surrogate_mean_squared_error_degradation(y: ndarray | list[float] | Series, y_pred: ndarray | list[float] | Series, y_surrogate: ndarray | list[float] | Series)

Calculate the difference between the mean squared error of the original model and the surrogate model.

Parameters

yArrayLike

The true target values.

y_predArrayLike

The predicted target values of the original model.

y_surrogateArrayLike

The predicted target values of the surrogate model.

Returns

float

The difference between the mean squared error of the original model and the surrogate model

Examples

>>> import numpy as np
>>> from holisticai.explainability.metrics.surrogate import (
...     surrogate_smape_difference,
... )
>>> y = np.array([1, 2, 3, 4, 5])
>>> y_pred = np.array([1.1, 2.2, 3.3, 4.4, 5.5])
>>> y_surrogate = np.array([1.2, 2.3, 3.4, 4.5, 5.6])
>>> surrogate_smape_difference(y, y_pred, y_surrogate)