[MRG]: Use coordinate_descent_gram when precompute is True | auto

doc/modules/linear_model.rst

@@ -184,7 +184,7 @@ for another implementation::
     >>> clf = linear_model.Lasso(alpha = 0.1)
     >>> clf.fit([[0, 0], [1, 1]], [0, 1])
     Lasso(alpha=0.1, copy_X=True, fit_intercept=True, max_iter=1000,
-       normalize=False, positive=False, precompute='auto', tol=0.0001,
+       normalize=False, positive=False, precompute=False, tol=0.0001,
        warm_start=False)
     >>> clf.predict([[1, 1]])
     array([ 0.8])

doc/tutorial/statistical_inference/model_selection.rst

@@ -196,7 +196,7 @@ automatically by cross-validation::
     >>> lasso.fit(X_diabetes, y_diabetes)
     LassoCV(alphas=None, copy_X=True, cv=None, eps=0.001, fit_intercept=True,
         max_iter=1000, n_alphas=100, n_jobs=1, normalize=False, positive=False,
-        precompute='auto', tol=0.0001, verbose=False)
+        precompute=False, tol=0.0001, verbose=False)
     >>> # The estimator chose automatically its lambda:
     >>> lasso.alpha_ # doctest: +ELLIPSIS
     0.01229...

doc/tutorial/statistical_inference/supervised_learning.rst

@@ -327,7 +327,7 @@ application of Occam's razor: `prefer simpler models`.
     >>> regr.alpha = best_alpha
     >>> regr.fit(diabetes_X_train, diabetes_y_train)
     Lasso(alpha=0.025118864315095794, copy_X=True, fit_intercept=True,
-       max_iter=1000, normalize=False, positive=False, precompute='auto',
+       max_iter=1000, normalize=False, positive=False, precompute=False,
        tol=0.0001, warm_start=False)
     >>> print(regr.coef_)
     [   0.         -212.43764548  517.19478111  313.77959962 -160.8303982    -0.

doc/whats_new.rst

@@ -291,6 +291,13 @@ API changes summary
      Previously it was broken for input of non-int dtype and the weighted
      array that was returned was wrong. By `Manoj Kumar`_.
 
+   - Change default value of precompute in :class:`ElasticNet`
+     :class:`ElasticNetCV`, :class:`Lasso` and :class:`LassoCV` from "auto"
+     to False. Setting precompute to "auto" was found to be slower since
+     the computation of the Gram matrix is computationally expensive and
+     outweighs the benefit of fitting the Gram.
+     By `Manoj Kumar`_.
+
 .. _changes_0_14:
 
 0.14

sklearn/linear_model/coordinate_descent.py

@@ -467,12 +467,19 @@ def enet_path(X, y, l1_ratio=0.5, eps=1e-3, n_alphas=100, alphas=None,
                 coef_, l1_reg, l2_reg, X.data, X.indices,
                 X.indptr, y, X_sparse_scaling,
                 max_iter, tol, positive)
-        elif not multi_output:
+        elif multi_output:
+            model = cd_fast.enet_coordinate_descent_multi_task(
+                coef_, l1_reg, l2_reg, X, y, max_iter, tol)
+        elif isinstance(precompute, np.ndarray):
+            model = cd_fast.enet_coordinate_descent_gram(
+                coef_, l1_reg, l2_reg, precompute, Xy, y, max_iter,
+                tol, positive)
+        elif precompute is False:
             model = cd_fast.enet_coordinate_descent(
                 coef_, l1_reg, l2_reg, X, y, max_iter, tol, positive)
         else:
-            model = cd_fast.enet_coordinate_descent_multi_task(
-                coef_, l1_reg, l2_reg, X, y, max_iter, tol)
+            raise ValueError("Precompute should be one of True, False, "
+                             "'auto' or array-like")
         coef_, dual_gap_, eps_ = model
         coefs[..., i] = coef_
         dual_gaps[i] = dual_gap_
@@ -615,7 +622,7 @@ class ElasticNet(LinearModel, RegressorMixin):
     path = staticmethod(enet_path)
 
     def __init__(self, alpha=1.0, l1_ratio=0.5, fit_intercept=True,
-                 normalize=False, precompute='auto', max_iter=1000,
+                 normalize=False, precompute=False, max_iter=1000,
                  copy_X=True, tol=1e-4, warm_start=False, positive=False):
         self.alpha = alpha
         self.l1_ratio = l1_ratio
@@ -802,7 +809,7 @@ class Lasso(ElasticNet):
     >>> clf = linear_model.Lasso(alpha=0.1)
     >>> clf.fit([[0,0], [1, 1], [2, 2]], [0, 1, 2])
     Lasso(alpha=0.1, copy_X=True, fit_intercept=True, max_iter=1000,
-       normalize=False, positive=False, precompute='auto', tol=0.0001,
+       normalize=False, positive=False, precompute=False, tol=0.0001,
        warm_start=False)
     >>> print(clf.coef_)
     [ 0.85  0.  ]
@@ -828,7 +835,7 @@ class Lasso(ElasticNet):
     path = staticmethod(enet_path)
 
     def __init__(self, alpha=1.0, fit_intercept=True, normalize=False,
-                 precompute='auto', copy_X=True, max_iter=1000,
+                 precompute=False, copy_X=True, max_iter=1000,
                  tol=1e-4, warm_start=False, positive=False):
         super(Lasso, self).__init__(
             alpha=alpha, l1_ratio=1.0, fit_intercept=fit_intercept,
@@ -889,7 +896,13 @@ def _path_residuals(X, y, train, test, path, path_params, alphas=None,
     y_test = y[test]
     fit_intercept = path_params['fit_intercept']
     normalize = path_params['normalize']
-    precompute = path_params['precompute']
+
+    if y.ndim == 1:
+        precompute = path_params['precompute']
+    else:
+        # No Gram variant of multi-task exists right now.
+        # Fall back to default enet_multitask
+        precompute = False
 
     X_train, y_train, X_mean, y_mean, X_std, precompute, Xy = \
         _pre_fit(X_train, y_train, None, precompute, normalize, fit_intercept,
@@ -1218,7 +1231,7 @@ class LassoCV(LinearModelCV, RegressorMixin):
     path = staticmethod(lasso_path)
 
     def __init__(self, eps=1e-3, n_alphas=100, alphas=None, fit_intercept=True,
-                 normalize=False, precompute='auto', max_iter=1000, tol=1e-4,
+                 normalize=False, precompute=False, max_iter=1000, tol=1e-4,
                  copy_X=True, cv=None, verbose=False, n_jobs=1,
                  positive=False):
         super(LassoCV, self).__init__(
@@ -1345,7 +1358,7 @@ class ElasticNetCV(LinearModelCV, RegressorMixin):
     path = staticmethod(enet_path)
 
     def __init__(self, l1_ratio=0.5, eps=1e-3, n_alphas=100, alphas=None,
-                 fit_intercept=True, normalize=False, precompute='auto',
+                 fit_intercept=True, normalize=False, precompute=False,
                  max_iter=1000, tol=1e-4, cv=None, copy_X=True,
                  verbose=0, n_jobs=1, positive=False):
         self.l1_ratio = l1_ratio
@@ -1638,11 +1651,6 @@ class MultiTaskElasticNetCV(LinearModelCV, RegressorMixin):
         List of alphas where to compute the models.
         If not provided, set automatically.
 
-    precompute : True | False | 'auto' | array-like
-        Whether to use a precomputed Gram matrix to speed up
-        calculations. If set to ``'auto'`` let us decide. The Gram
-        matrix can also be passed as argument.
-
     n_alphas : int, optional
         Number of alphas along the regularization path
 
@@ -1716,8 +1724,7 @@ class MultiTaskElasticNetCV(LinearModelCV, RegressorMixin):
     ... #doctest: +NORMALIZE_WHITESPACE
     MultiTaskElasticNetCV(alphas=None, copy_X=True, cv=None, eps=0.001,
            fit_intercept=True, l1_ratio=0.5, max_iter=1000, n_alphas=100,
-           n_jobs=1, normalize=False, precompute='auto', tol=0.0001,
-           verbose=0)
+           n_jobs=1, normalize=False, tol=0.0001, verbose=0)
     >>> print(clf.coef_)
     [[ 0.52875032  0.46958558]
      [ 0.52875032  0.46958558]]
@@ -1740,7 +1747,7 @@ class MultiTaskElasticNetCV(LinearModelCV, RegressorMixin):
     path = staticmethod(enet_path)
 
     def __init__(self, l1_ratio=0.5, eps=1e-3, n_alphas=100, alphas=None,
-                 fit_intercept=True, normalize=False, precompute='auto',
+                 fit_intercept=True, normalize=False,
                  max_iter=1000, tol=1e-4, cv=None, copy_X=True,
                  verbose=0, n_jobs=1):
         self.l1_ratio = l1_ratio
@@ -1749,7 +1756,6 @@ def __init__(self, l1_ratio=0.5, eps=1e-3, n_alphas=100, alphas=None,
         self.alphas = alphas
         self.fit_intercept = fit_intercept
         self.normalize = normalize
-        self.precompute = precompute
         self.max_iter = max_iter
         self.tol = tol
         self.cv = cv
@@ -1781,11 +1787,6 @@ class MultiTaskLassoCV(LinearModelCV, RegressorMixin):
         List of alphas where to compute the models.
         If not provided, set automaticlly.
 
-    precompute : True | False | 'auto' | array-like
-        Whether to use a precomputed Gram matrix to speed up
-        calculations. If set to ``'auto'`` let us decide. The Gram
-        matrix can also be passed as argument.
-
     n_alphas : int, optional
         Number of alphas along the regularization path
 
@@ -1856,10 +1857,10 @@ class MultiTaskLassoCV(LinearModelCV, RegressorMixin):
     path = staticmethod(lasso_path)
 
     def __init__(self, eps=1e-3, n_alphas=100, alphas=None, fit_intercept=True,
-                 normalize=False, precompute='auto', max_iter=1000, tol=1e-4,
-                 copy_X=True, cv=None, verbose=False, n_jobs=1):
+                 normalize=False, max_iter=1000, tol=1e-4, copy_X=True,
+                 cv=None, verbose=False, n_jobs=1):
         super(MultiTaskLassoCV, self).__init__(
             eps=eps, n_alphas=n_alphas, alphas=alphas,
             fit_intercept=fit_intercept, normalize=normalize,
-            precompute=precompute, max_iter=max_iter, tol=tol, copy_X=copy_X,
+            max_iter=max_iter, tol=tol, copy_X=copy_X,
             cv=cv, verbose=verbose, n_jobs=n_jobs)

sklearn/linear_model/tests/test_coordinate_descent.py

@@ -460,6 +460,13 @@ def test_sparse_input_dtype_enet_and_lassocv():
     assert_almost_equal(clf.coef_, clf1.coef_, decimal=6)
 
 
+def test_precompute_invalid_argument():
+    X, y, _, _ = build_dataset()
+    for clf in [ElasticNetCV(precompute="invalid"),
+                LassoCV(precompute="invalid")]:
+        assert_raises(ValueError, clf.fit, X, y)
+
+
 if __name__ == '__main__':
     import nose
     nose.runmodule()