Store the folds separately #21664
-
|
In the |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 3 replies
-
|
Currently not but with the work in #21211, it looks like having a parameter If we go on this direction for the plotting API, then we might want to add |
Beta Was this translation helpful? Give feedback.
-
|
@glemaitre kfold = KFold(n_splits=cv, shuffle=False,
random_state=random_state)
index = list(kfold.split(X, y))[0] |
Beta Was this translation helpful? Give feedback.
-
|
Yep but there is a risk with shuffling if not |
Beta Was this translation helpful? Give feedback.
-
@glemaitre Thank you so much |
Beta Was this translation helpful? Give feedback.
Currently not but with the work in #21211, it looks like having a parameter
return_indiceswould be handy: https://github.com/scikit-learn/scikit-learn/pull/21211/files#diff-24fbe29b336ea0ad7ef67d54c362bfebd086d2367337526fb99dae4583891c2dIf we go on this direction for the plotting API, then we might want to add
return_indicesandreturn_predictionsto thecross_validatefunction.