Merge pull request scikit-learn#6426 from yenchenlin1994/remove-redundant-typo

[MRG] DOC Remove redundant words in sklearn

Author: GaelVaroquaux

doc/faq.rst

@@ -235,7 +235,7 @@ anymore. The version of joblib shipped with scikit-learn automatically uses
 that setting by default (under Python 3.4 and later).
 
 If you have custom code that uses ``multiprocessing`` directly instead of using
-it via joblib you can enable the the 'forkserver' mode globally for your
+it via joblib you can enable the 'forkserver' mode globally for your
 program: Insert the following instructions in your main script::
 
  import multiprocessing

doc/modules/biclustering.rst

@@ -140,7 +140,7 @@ are used to form the matrix :math:`Z`:
  C^{-1/2} V
  \end{bmatrix}
 
-where the the columns of :math:`U` are :math:`u_2, \dots, u_{\ell +
+where the columns of :math:`U` are :math:`u_2, \dots, u_{\ell +
 1}`, and similarly for :math:`V`.
 
 Then the rows of :math:`Z` are clustered using :ref:`k-means
@@ -174,7 +174,7 @@ The :class:`SpectralBiclustering` algorithm assumes that the input
 data matrix has a hidden checkerboard structure. The rows and columns
 of a matrix with this structure may be partitioned so that the entries
 of any bicluster in the Cartesian product of row clusters and column
-clusters is are approximately constant. For instance, if there are two
+clusters are approximately constant. For instance, if there are two
 row partitions and three column partitions, each row will belong to
 three biclusters, and each column will belong to two biclusters.
 

doc/modules/covariance.rst

@@ -280,7 +280,7 @@ empirical covariance matrix is then rescaled to compensate the
 performed selection of observations ("consistency step"). Having
 computed the Minimum Covariance Determinant estimator, one can give
 weights to observations according to their Mahalanobis distance,
-leading the a reweighted estimate of the covariance matrix of the data
+leading to a reweighted estimate of the covariance matrix of the data
 set ("reweighting step").
 
 Rousseeuw and Van Driessen [4] developed the FastMCD algorithm in order

doc/modules/cross_validation.rst

@@ -465,7 +465,7 @@ Here is a usage example::
 
 :class:`ShuffleSplit` is thus a good alternative to :class:`KFold` cross
 validation that allows a finer control on the number of iterations and
-the proportion of samples in on each side of the train / test split.
+the proportion of samples on each side of the train / test split.
 
 
 Label-Shuffle-Split

doc/modules/ensemble.rst

@@ -679,7 +679,7 @@ the contribution of each weak learner by a factor :math:`\nu`:
  F_m(x) = F_{m-1}(x) + \nu \gamma_m h_m(x)
 
 The parameter :math:`\nu` is also called the **learning rate** because
-it scales the step length the the gradient descent procedure; it can
+it scales the step length the gradient descent procedure; it can
 be set via the ``learning_rate`` parameter.
 
 The parameter ``learning_rate`` strongly interacts with the parameter

doc/modules/pipeline.rst

@@ -134,7 +134,7 @@ create complex models.
 
 (A :class:`FeatureUnion` has no way of checking whether two transformers
 might produce identical features. It only produces a union when the
-feature sets are disjoint, and making sure they are is the caller's
+feature sets are disjoint, and making sure they are the caller's
 responsibility.)
 
 

doc/modules/svm.rst

@@ -181,7 +181,7 @@ for these classifiers.
 
 This might be made more clear by an example:
 
-Consider a three class problem with with class 0 having three support vectors
+Consider a three class problem with class 0 having three support vectors
 :math:`v^{0}_0, v^{1}_0, v^{2}_0` and class 1 and 2 having two support vectors
 :math:`v^{0}_1, v^{1}_1` and :math:`v^{0}_2, v^{1}_2` respectively. For each
 support vector :math:`v^{j}_i`, there are two dual coefficients. Let's call

doc/tutorial/text_analytics/skeletons/exercise_01_language_train_model.py

@@ -28,7 +28,7 @@
  dataset.data, dataset.target, test_size=0.5)
 
 
-# TASK: Build a an vectorizer that splits strings into sequence of 1 to 3
+# TASK: Build a vectorizer that splits strings into sequence of 1 to 3
 # characters instead of word tokens
 
 # TASK: Build a vectorizer / classifier pipeline using the previous analyzer

doc/tutorial/text_analytics/solutions/exercise_01_language_train_model.py

@@ -28,7 +28,7 @@
  dataset.data, dataset.target, test_size=0.5)
 
 
-# TASK: Build a an vectorizer that splits strings into sequence of 1 to 3
+# TASK: Build a vectorizer that splits strings into sequence of 1 to 3
 # characters instead of word tokens
 vectorizer = TfidfVectorizer(ngram_range=(1, 3), analyzer='char',
  use_idf=False)

doc/whats_new.rst

@@ -2812,7 +2812,7 @@ Other changes
 
  - :class:`svm.SVC` members ``coef_`` and ``intercept_`` changed sign for
  consistency with ``decision_function``; for ``kernel==linear``,
- ``coef_`` was fixed in the the one-vs-one case, by `Andreas Müller`_.
+ ``coef_`` was fixed in the one-vs-one case, by `Andreas Müller`_.
 
  - Performance improvements to efficient leave-one-out cross-validated
  Ridge regression, esp. for the ``n_samples > n_features`` case, in
@@ -2993,7 +2993,7 @@ Changelog
 
  - Fixed a bug in the RFE module by `Gilles Louppe`_ (issue #378).
 
- - Fixed a memory leak in in :ref:`svm` module by `Brian Holt`_ (issue #367).
+ - Fixed a memory leak in :ref:`svm` module by `Brian Holt`_ (issue #367).
 
  - Faster tests by `Fabian Pedregosa`_ and others.