Add decision_function to ElasticNet.

mblondel · mblondel · commit 68236ab14c5f · 2012-01-10T16:22:05.000+09:00
This allows to plug ElasticNet and Lasso into the multiclass estimators.
diff --git a/sklearn/linear_model/base.py b/sklearn/linear_model/base.py
@@ -41,8 +41,8 @@
 class LinearModel(BaseEstimator, RegressorMixin):
     """Base class for Linear Models"""
 
-    def predict(self, X):
-        """Predict using the linear model
+    def decision_function(self, X):
+        """Decision function of the linear model
 
         Parameters
         ----------
@@ -56,6 +56,20 @@ def predict(self, X):
         X = safe_asarray(X)
         return safe_sparse_dot(X, self.coef_.T) + self.intercept_
 
+    def predict(self, X):
+        """Predict using the linear model
+
+        Parameters
+        ----------
+        X : numpy array of shape [n_samples, n_features]
+
+        Returns
+        -------
+        C : array, shape = [n_samples]
+            Returns predicted values.
+        """
+        return self.decision_function(X)
+
     @staticmethod
     def _center_data(X, y, fit_intercept, normalize=False, copy=True):
         """
@@ -250,7 +264,7 @@ def _validate_sample_weight(self, sample_weight, n_samples):
 
     def _set_coef(self, coef_):
         """Make sure that coef_ is fortran-style and 2d.
-        
+
         Fortran-style memory layout is needed to ensure that computing
         the dot product between input ``X`` and ``coef_`` does not trigger
         a memory copy.
diff --git a/sklearn/linear_model/ridge.py b/sklearn/linear_model/ridge.py
@@ -284,7 +284,7 @@ def fit(self, X, y, solver='auto'):
         return self
 
     def decision_function(self, X):
-        return Ridge.predict(self, X)
+        return Ridge.decision_function(self, X)
 
     def predict(self, X):
         """Predict target values according to the fitted model.
@@ -569,7 +569,7 @@ def fit(self, X, y, sample_weight=1.0, class_weight=None):
         return self
 
     def decision_function(self, X):
-        return RidgeCV.predict(self, X)
+        return RidgeCV.decision_function(self, X)
 
     def predict(self, X):
         """Predict target values according to the fitted model.
diff --git a/sklearn/linear_model/sparse/coordinate_descent.py b/sklearn/linear_model/sparse/coordinate_descent.py
@@ -105,8 +105,8 @@ def fit(self, X, y):
         # return self for chaining fit and predict calls
         return self
 
-    def predict(self, X):
-        """Predict using the linear model
+    def decision_function(self, X):
+        """Decision function of the linear model
 
         Parameters
         ----------
