scikit-learn · glemaitre · Oct 1, 2021 · Oct 1, 2021 · Oct 1, 2021 · Oct 1, 2021
diff --git a/doc/modules/cross_validation.rst b/doc/modules/cross_validation.rst
@@ -208,8 +208,8 @@ two ways:
 - It allows specifying multiple metrics for evaluation.
 
 - It returns a dict containing fit-times, score-times
-  (and optionally training scores as well as fitted estimators) in
-  addition to the test score.
+  (and optionally training scores, fitted estimators, train-test split indices)
+  in addition to the test score.
 
 For single metric evaluation, where the scoring parameter is a string,
 callable or None, the keys will be - ``['test_score', 'fit_time', 'score_time']``
@@ -220,10 +220,10 @@ following keys -
 
 ``return_train_score`` is set to ``False`` by default to save computation time.
 To evaluate the scores on the training set as well you need to set it to
-``True``.
-
-You may also retain the estimator fitted on each training set by setting
-``return_estimator=True``.
+``True``. You may also retain the estimator fitted on each training set by
+setting ``return_estimator=True``. On the same line of thought, you may set
+`return_indices=True` to retain the training and testing indices used to split
+the dataset into train and test sets for each cv split.
 
 The multiple metrics can be specified either as a list, tuple or set of
 predefined scorer names::

diff --git a/doc/whats_new/v1.1.rst b/doc/whats_new/v1.1.rst
@@ -38,13 +38,29 @@ Changelog
     :pr:`123456` by :user:`Joe Bloggs <joeongithub>`.
     where 123456 is the *pull request* number, not the issue number.
 
+:mod:`sklearn.calibration`
+..........................
+
+- |Feature| :class:`calibration.CalibrationDisplay` has a new method
+  :meth:`~calibration.CalibrationDisplay.from_cv_results` allowing to plot
+  calibration curves together with standard deviation uncertainty using
+  cross-validation results.
+  :pr:`21211` by :user:`Guillaume Lemaitre <glemaitre>`.
+
 :mod:`sklearn.linear_model`
 ...........................
 
 - |Fix| :class:`linear_model.LogisticRegression` now raises a better error
   message when the solver does not support sparse matrices with int64 indices.
   :pr:`21093` by `Tom Dupre la Tour`_.
 
+:mod:`sklearn.model_selection`
+..............................
+
+- |Enhancement| :func:`model_selection.cross_validate` accepts a new parameter
+  `return_indices` to return the train-test indices of each cv split.
+  :pr:`21211` by :user:`Guillaume Lemaitre <glemaitre>`.
+
 :mod:`sklearn.utils`
 ....................