il me semble que c’est une bonne reponse.
Q3)
g) the 5 trained estimators are returned
d’après ce paragraphe:
Note that by default the cross_validate function discards the 5 models that were trained on the different overlapping subset of the dataset.
N’est ce pas les amis?
Note that by default the
cross_validatefunction discards the 5 models that were trained on the different overlapping subset of the dataset.
The fact that the models are discarded (by default) means that they are not returned.
Note that it’s possible to call cross_validate(model, X, y, cv=5, return_estimator=True) to avoid discarding and return the models instead.