-
-
Notifications
You must be signed in to change notification settings - Fork 26.5k
[MRG] Allow nd array for CalibratedClassifierCV #13485
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[MRG] Allow nd array for CalibratedClassifierCV #13485
Conversation
|
Is there a reason why scikit-learn/sklearn/calibration.py Line 31 in 7684298
|
|
Probably no reason. MetaEstimatorMixin did very little until recently. Now
it only does little.
|
|
Please add a non-regression test |
|
@wdevazelhes You might be interested in this PR (by me) which should work with nd-array input (and also supports multi-label probability calibration). It includes tests which you can use if you'd like, to verify that input does not need to be 2d. However, it doesn't include any tests for arrays with more than two dimensions. |
|
@connorbrinton Thanks for the reference ! Great, yes indeed your PR would fix my problem with nd-array :) |
|
Please merge latest master |
Done |
|
Please add an |
done |
Reference Issues/PRs
Fixes #13077
What does this implement/fix? Explain your changes.
This allows n-dimensional arrays for
CalibratedClassifierCV(for instance,GridSearchCVallows for all kind of arrays, since it's aMetaEstimatorand more precise tests are made by the encapsulated estimator itself). See #13077 for more details.