DOC sparse_coef_ documentation for MultiTaskElasticNet and MultiTaskL…

…asso (scikit-learn#17884)

sklearn/linear_model/_coordinate_descent.py

@@ -651,9 +651,9 @@ class ElasticNet(MultiOutputMixin, RegressorMixin, LinearModel):
  coef_ : ndarray of shape (n_features,) or (n_targets, n_features)
  parameter vector (w in the cost function formula)
 
- sparse_coef_ : sparse matrix of shape (n_features, 1) or \
- (n_targets, n_features)
- ``sparse_coef_`` is a readonly property derived from ``coef_``
+ sparse_coef_ : sparse matrix of shape (n_features,) or \
+ (n_tasks, n_features)
+ sparse representation of the `coef_`.
 
  intercept_ : float or ndarray of shape (n_targets,)
  independent term in decision function.
@@ -1840,6 +1840,10 @@ class MultiTaskElasticNet(Lasso):
  eps_ : float
  The tolerance scaled scaled by the variance of the target `y`.
 
+ sparse_coef_ : sparse matrix of shape (n_features,) or \
+ (n_tasks, n_features)
+ sparse representation of the `coef_`.
+
  Examples
  --------
  >>> from sklearn import linear_model
@@ -2035,6 +2039,10 @@ class MultiTaskLasso(MultiTaskElasticNet):
  eps_ : float
  The tolerance scaled scaled by the variance of the target `y`.
 
+ sparse_coef_ : sparse matrix of shape (n_features,) or \
+ (n_tasks, n_features)
+ sparse representation of the `coef_`.
+
  Examples
  --------
  >>> from sklearn import linear_model