According to the formula in official page https://scikit-learn.org/stable/modules/model_evaluation.html#explained-variance-score, in order to calculate following EVS for the data set:
y_true = [1, 2, 3, 4, 5] y_pred = [6, 7, 8, 9, 10]
Manually: evs = 1 - var(y_true - y_pred)/var(y_true) = -11.5
Using code: evs = 1
from sklearn.metrics import explained_variance_score
y_true = [1, 2, 3, 4, 5]
y_pred = [6, 7, 8, 9, 10]
explained_variance = explained_variance_score(y_true, y_pred)
Why is the result different?