Merge pull request scikit-learn#5157 from amueller/ridge_intercept_docs

[MRG] DOC mention intercept_ attribute in ridge docstring.

Author: agramfort

sklearn/linear_model/ridge.py

@@ -456,6 +456,10 @@ class Ridge(_BaseRidge, RegressorMixin):
     coef_ : array, shape = [n_features] or [n_targets, n_features]
         Weight vector(s).
 
+    intercept_ : float | array, shape = (n_targets,)
+        Independent term in decision function. Set to 0.0 if
+        ``fit_intercept = False``.
+
     See also
     --------
     RidgeClassifier, RidgeCV, KernelRidge
@@ -554,6 +558,10 @@ class RidgeClassifier(LinearClassifierMixin, _BaseRidge):
     coef_ : array, shape = [n_features] or [n_classes, n_features]
         Weight vector(s).
 
+    intercept_ : float | array, shape = (n_targets,)
+        Independent term in decision function. Set to 0.0 if
+        ``fit_intercept = False``.
+
     See also
     --------
     Ridge, RidgeClassifierCV
@@ -875,7 +883,7 @@ def fit(self, X, y, sample_weight=None):
                 raise ValueError("cv!=None and store_cv_values=True "
                                  " are incompatible")
             parameters = {'alpha': self.alphas}
-            fit_params = {'sample_weight' : sample_weight}
+            fit_params = {'sample_weight': sample_weight}
             gs = GridSearchCV(Ridge(fit_intercept=self.fit_intercept),
                               parameters, fit_params=fit_params, cv=self.cv)
             gs.fit(X, y)
@@ -957,13 +965,13 @@ class RidgeCV(_BaseRidgeCV, RegressorMixin):
     coef_ : array, shape = [n_features] or [n_targets, n_features]
         Weight vector(s).
 
-    alpha_ : float
-        Estimated regularization parameter.
-
     intercept_ : float | array, shape = (n_targets,)
         Independent term in decision function. Set to 0.0 if
         ``fit_intercept = False``.
 
+    alpha_ : float
+        Estimated regularization parameter.
+
     See also
     --------
     Ridge: Ridge regression
@@ -1028,6 +1036,10 @@ class RidgeClassifierCV(LinearClassifierMixin, _BaseRidgeCV):
     coef_ : array, shape = [n_features] or [n_targets, n_features]
         Weight vector(s).
 
+    intercept_ : float | array, shape = (n_targets,)
+        Independent term in decision function. Set to 0.0 if
+        ``fit_intercept = False``.
+
     alpha_ : float
         Estimated regularization parameter