DOC: restructure docstring of ElasticNet.

Fabian Pedregosa · Fabian Pedregosa · commit c0ed2affa15f · 2012-01-10T17:30:37.000+01:00
If found it confusing that remarks about rho parameter where scattered
in heading and notes, moved everything related to the objective
function into the main part. No new content was added.
diff --git a/sklearn/linear_model/coordinate_descent.py b/sklearn/linear_model/coordinate_descent.py
@@ -20,8 +20,24 @@
 class ElasticNet(LinearModel):
     """Linear Model trained with L1 and L2 prior as regularizer
 
-    rho = 1 is the lasso penalty. Currently, rho <= 0.01 is not
-    reliable, unless you supply your own sequence of alpha.
+    Minimizes the objective function::
+
+            1 / (2 * n_samples) * ||y - Xw||^2_2 +
+            + alpha * rho * ||w||_1 + 0.5 * alpha * (1 - rho) * ||w||^2_2
+
+    If you are interested in controlling the L1 and L2 penalty
+    separately, keep in mind that this is equivalent to::
+
+            a * L1 + b * L2
+
+    where::
+
+            alpha = a + b and rho = a / (a + b)
+
+    The parameter rho corresponds to alpha in the glmnet R package while
+    alpha corresponds to the lambda parameter in glmnet. Specifically, rho =
+    1 is the lasso penalty. Currently, rho <= 0.01 is not reliable, unless
+    you supply your own sequence of alpha.
 
     Parameters
     ----------
@@ -63,23 +79,6 @@ class ElasticNet(LinearModel):
     -----
     To avoid unnecessary memory duplication the X argument of the fit method
     should be directly passed as a fortran contiguous numpy array.
-
-    The parameter rho corresponds to alpha in the glmnet R package
-    while alpha corresponds to the lambda parameter in glmnet.
-    More specifically, the objective function is::
-
-        1 / (2 * n_samples) * ||y - Xw||^2_2 +
-        + alpha * rho * ||w||_1 + 0.5 * alpha * (1 - rho) * ||w||^2_2
-
-    If you are interested in controlling the L1 and L2 penalty
-    separately, keep in mind that this is equivalent to::
-
-        a * L1 + b * L2
-
-    for::
-
-        alpha = a + b and rho = a / (a + b)
-
     """
     def __init__(self, alpha=1.0, rho=0.5, fit_intercept=True,
                  normalize=False, precompute='auto', max_iter=1000,
