DOC parameter is called "sample_weight", not "sample_weights" (scikit…

…-learn#7782)
neurodata · Oct 29, 2016 · 4c8f12d · 4c8f12d
1 parent e841996
commit 4c8f12d
Showing 1 changed file with 3 additions and 3 deletions.
diff --git a/doc/whats_new.rst b/doc/whats_new.rst
@@ -356,7 +356,7 @@ Trees and ensembles
  by dynamically generating attribute ``estimators_samples_`` only when it is
  needed. By `David Staub`_.
 
- - Added ``n_jobs`` and ``sample_weights`` parameters for
+ - Added ``n_jobs`` and ``sample_weight`` parameters for
  :class:`ensemble.VotingClassifier` to fit underlying estimators in parallel.
  :issue:`5805` by `Ibraim Ganiev`_.
 
@@ -366,7 +366,7 @@ Linear, kernelized and related models
  available in the multinomial case. :issue:`5251` by `Tom Dupre la Tour`_.
 
  - :class:`linear_model.RANSACRegressor`, :class:`svm.LinearSVC` and
- :class:`svm.LinearSVR` now support ``sample_weights``.
+ :class:`svm.LinearSVR` now support ``sample_weight``.
  By `Imaculate`_.
 
  - Add parameter ``loss`` to :class:`linear_model.RANSACRegressor` to measure the
@@ -882,7 +882,7 @@ Enhancements
  - :class:`cluster.mean_shift_.MeanShift` now supports parallel execution,
  as implemented in the ``mean_shift`` function. By `Martino Sorbaro`_.
 
- - :class:`naive_bayes.GaussianNB` now supports fitting with ``sample_weights``.
+ - :class:`naive_bayes.GaussianNB` now supports fitting with ``sample_weight``.
  By `Jan Hendrik Metzen`_.
 
  - :class:`dummy.DummyClassifier` now supports a prior fitting strategy.