quantlab.model.evaluation.r2

quantlab.model.evaluation.r2(pred, target)

Return the coefficient of determination 1 - SS_res / SS_tot.

NaN is returned when fewer than two cells are usable or when the target is constant over the usable cells (SS_tot == 0), because R2 is undefined in both cases.

Parameters:

predarray_like

Predictions, any shape.

targetarray_like

Realised values, the same shape as pred.

Examples

>>> r2([1.0, 2.0, 3.0, 4.0], [1.0, 2.0, 3.0, 6.0])
0.7142857142857143
>>> r2([1.0, 2.0], [3.0, 3.0])
nan