scikit-learn · agramfort · Jul 11, 2017 · Jun 6, 2017 · Jun 6, 2017 · Jun 7, 2017
diff --git a/.travis.yml b/.travis.yml
@@ -49,6 +49,7 @@ matrix:
     - env: RUN_FLAKE8="true" SKIP_TESTS="true"
            DISTRIB="conda" PYTHON_VERSION="3.5" INSTALL_MKL="true"
            NUMPY_VERSION="1.12.1" SCIPY_VERSION="0.19.0" CYTHON_VERSION="0.23.5"
+           TEST_DOCSTRINGS="true"
     # This environment tests scikit-learn against numpy and scipy master
     # installed from their CI wheels in a virtualenv with the Python
     # interpreter provided by travis.

diff --git a/build_tools/travis/install.sh b/build_tools/travis/install.sh
@@ -85,6 +85,10 @@ if [[ "$COVERAGE" == "true" ]]; then
     pip install coverage codecov
 fi
 
+if [[ "$TEST_DOCSTRINGS" == "true" ]]; then
+    pip install sphinx numpydoc  # numpydoc requires sphinx
+fi
+
 if [[ "$SKIP_TESTS" == "true" ]]; then
     echo "No need to build scikit-learn when not running the tests"
 else

diff --git a/sklearn/base.py b/sklearn/base.py
@@ -428,6 +428,11 @@ def get_indices(self, i):
 
         Only works if ``rows_`` and ``columns_`` attributes exist.
 
+        Parameters
+        ----------
+        i : int
+            The index of the cluster.
+
         Returns
         -------
         row_ind : np.array, dtype=np.intp
@@ -443,6 +448,11 @@ def get_indices(self, i):
     def get_shape(self, i):
         """Shape of the i'th bicluster.
 
+        Parameters
+        ----------
+        i : int
+            The index of the cluster.
+
         Returns
         -------
         shape : (int, int)
@@ -454,9 +464,22 @@ def get_shape(self, i):
     def get_submatrix(self, i, data):
         """Returns the submatrix corresponding to bicluster `i`.
 
+        Parameters
+        ----------
+        i : int
+            The index of the cluster.
+        data : array
+            The data.
+
+        Returns
+        -------
+        submatrix : array
+            The submatrix corresponding to bicluster i.
+
+        Notes
+        -----
         Works with sparse matrices. Only works if ``rows_`` and
         ``columns_`` attributes exist.
-
         """
         from .utils.validation import check_array
         data = check_array(data, accept_sparse='csr')
@@ -525,10 +548,33 @@ class MetaEstimatorMixin(object):
 ###############################################################################
 
 def is_classifier(estimator):
-    """Returns True if the given estimator is (probably) a classifier."""
+    """Returns True if the given estimator is (probably) a classifier.
+
+    Parameters
+    ----------
+    estimator : object
+        Estimator object to test.
+
+    Returns
+    -------
+    out : bool
+        True if estimator is a classifier and False otherwise.
+    """
     return getattr(estimator, "_estimator_type", None) == "classifier"
 
 
 def is_regressor(estimator):
-    """Returns True if the given estimator is (probably) a regressor."""
+    """Returns True if the given estimator is (probably) a regressor.
+
+
+    Parameters
+    ----------
+    estimator : object
+        Estimator object to test.
+
+    Returns
+    -------
+    out : bool
+        True if estimator is a regressor and False otherwise.
+    """
     return getattr(estimator, "_estimator_type", None) == "regressor"
diff --git a/sklearn/cluster/affinity_propagation_.py b/sklearn/cluster/affinity_propagation_.py
@@ -199,13 +199,13 @@ class AffinityPropagation(BaseEstimator, ClusterMixin):
     damping : float, optional, default: 0.5
         Damping factor between 0.5 and 1.
 
+    max_iter : int, optional, default: 200
+        Maximum number of iterations.
+
     convergence_iter : int, optional, default: 15
         Number of iterations with no change in the number
         of estimated clusters that stops the convergence.
 
-    max_iter : int, optional, default: 200
-        Maximum number of iterations.
-
     copy : boolean, optional, default: True
         Make a copy of input data.
 

diff --git a/sklearn/cluster/hierarchical.py b/sklearn/cluster/hierarchical.py
@@ -312,6 +312,9 @@ def linkage_tree(X, connectivity=None, n_components=None,
         be symmetric and only the upper triangular half is used.
         Default is None, i.e, the Ward algorithm is unstructured.
 
+    n_components : int (optional)
+        The number of connected components in the graph.
+
     n_clusters : int (optional)
         Stop early the construction of the tree at n_clusters. This is
         useful to decrease computation time if the number of clusters is
@@ -596,14 +599,6 @@ class AgglomerativeClustering(BaseEstimator, ClusterMixin):
     n_clusters : int, default=2
         The number of clusters to find.
 
-    connectivity : array-like or callable, optional
-        Connectivity matrix. Defines for each sample the neighboring
-        samples following a given structure of the data.
-        This can be a connectivity matrix itself or a callable that transforms
-        the data into a connectivity matrix, such as derived from
-        kneighbors_graph. Default is None, i.e, the
-        hierarchical clustering algorithm is unstructured.
-
     affinity : string or callable, default: "euclidean"
         Metric used to compute the linkage. Can be "euclidean", "l1", "l2",
         "manhattan", "cosine", or 'precomputed'.
@@ -615,6 +610,14 @@ class AgglomerativeClustering(BaseEstimator, ClusterMixin):
         By default, no caching is done. If a string is given, it is the
         path to the caching directory.
 
+    connectivity : array-like or callable, optional
+        Connectivity matrix. Defines for each sample the neighboring
+        samples following a given structure of the data.
+        This can be a connectivity matrix itself or a callable that transforms
+        the data into a connectivity matrix, such as derived from
+        kneighbors_graph. Default is None, i.e, the
+        hierarchical clustering algorithm is unstructured.
+
     compute_full_tree : bool or 'auto' (optional)
         Stop early the construction of the tree at n_clusters. This is
         useful to decrease computation time if the number of clusters is
@@ -766,14 +769,6 @@ class FeatureAgglomeration(AgglomerativeClustering, AgglomerationTransform):
     n_clusters : int, default 2
         The number of clusters to find.
 
-    connectivity : array-like or callable, optional
-        Connectivity matrix. Defines for each feature the neighboring
-        features following a given structure of the data.
-        This can be a connectivity matrix itself or a callable that transforms
-        the data into a connectivity matrix, such as derived from
-        kneighbors_graph. Default is None, i.e, the
-        hierarchical clustering algorithm is unstructured.
-
     affinity : string or callable, default "euclidean"
         Metric used to compute the linkage. Can be "euclidean", "l1", "l2",
         "manhattan", "cosine", or 'precomputed'.
@@ -785,6 +780,14 @@ class FeatureAgglomeration(AgglomerativeClustering, AgglomerationTransform):
         By default, no caching is done. If a string is given, it is the
         path to the caching directory.
 
+    connectivity : array-like or callable, optional
+        Connectivity matrix. Defines for each feature the neighboring
+        features following a given structure of the data.
+        This can be a connectivity matrix itself or a callable that transforms
+        the data into a connectivity matrix, such as derived from
+        kneighbors_graph. Default is None, i.e, the
+        hierarchical clustering algorithm is unstructured.
+
     compute_full_tree : bool or 'auto', optional, default "auto"
         Stop early the construction of the tree at n_clusters. This is
         useful to decrease computation time if the number of clusters is