MAINT validate parameter in Ridge #23563
Conversation
There was a problem hiding this comment.
Thanks for the PR @eliaschiavon. Here are some suggestions.
The next step would to call self._validate_params in fit in Ridge and RidgeClassifier which inherit from this base class. Then check if previous validations are now obsolete and can be removed.
|
Thank you very much for the feedback, Jeremie. I'm gonna implement all the things the next week since I'm away during the weekend. |
…ter constraints _BaseRidge
-removed Ridge and RidgeClassifier from list of PARAM_VALIDATION_ESTIMATORS_TO_IGNORE
normalize in parameter constraints in BaseRidge is now an Hidden StrOptions
|
All the things should be ok for Ridge and RidgeClassifier. The only thing failing in the test is the implementation of the normalize parameter as Hidden, as you suggested in the comments above. |
|
@eliaschiavon I directly pushed some changes to take into account the recent improvements in the validation mechanism. |
sklearn/linear_model/_ridge.py
Outdated
| @@ -637,7 +641,7 @@ def _ridge_regression( | |||
|
|
|||
| # Some callers of this method might pass alpha as single | |||
| # element array which already has been validated. | |||
| if alpha is not None and not isinstance(alpha, (np.ndarray, tuple)): | |||
| if alpha is not None and not isinstance(alpha, (np.ndarray)): | |||
There was a problem hiding this comment.
we can probably remove the paranthesis around np.ndarray.
We are sure that we don't have any case where the value in the tuple have been validated but not transformed to a NumPy array.
|
Thank you for your work @jeremiedbb and @glemaitre ! |
Co-authored-by: jeremie du boisberranger <jeremiedbb@yahoo.fr> Co-authored-by: Jérémie du Boisberranger <34657725+jeremiedbb@users.noreply.github.com>
Reference Issues/PRs
Towards #23462
What does this implement/fix? Explain your changes.
Use of _validate_params for the Ridge family of algorithms