how to calculate classification accuracy sklearn code example
Example 1: accuracy score sklearn syntax
>>> from sklearn.metrics import accuracy_score
>>> y_pred = [0, 2, 1, 3]
>>> y_true = [0, 1, 2, 3]
>>> accuracy_score(y_true, y_pred)
0.5
>>> accuracy_score(y_true, y_pred, normalize=False)
2
Example 2: sklearn.metrics accuracy_score
// syntax:
// - sklearn.metrics.accuracy_score(y_true, y_pred, *, normalize=True, sample_weight=None)