new test passes

btel · btel · commit 0a2cfd7bc0ed · 2019-03-01T17:08:12.000+01:00
diff --git a/sklearn/linear_model/ridge.py b/sklearn/linear_model/ridge.py
@@ -370,12 +370,12 @@ def _ridge_regression(X, y, alpha, sample_weight=None, solver='auto',
     has_sw = sample_weight is not None
 
     def _select_auto_mode():
-        if has_sw:
-            # this may be changed, since all solver support sample_weights
-            return "cholesky"
         if return_intercept:
             # only sag and saga support fitting intercept directly
             return "sag"
+        if has_sw:
+            # this should be changed since all solvers support sample_weights
+            return "cholesky"
         if sparse.issparse(X):
             return "sparse_cg"
         return "cholesky"
diff --git a/sklearn/linear_model/tests/test_ridge.py b/sklearn/linear_model/tests/test_ridge.py
@@ -836,34 +836,26 @@ def test_ridge_fit_intercept_sparse():
         assert_array_almost_equal(dense.coef_, sparse.coef_)
 
 
-@pytest.mark.parametrize('return_intercept', [True, False])
-@pytest.mark.parametrize('sample_weight', [None, np.random.rand(1000)])
+@pytest.mark.parametrize('return_intercept', [False, True])
+@pytest.mark.parametrize('sample_weight', [None, np.ones(1000)])
 @pytest.mark.parametrize('arr_type', [np.array, sp.csr_matrix])
 def test_ridge_check_auto_modes(return_intercept, sample_weight, arr_type):
-    X, y = make_regression(n_samples=1000, n_features=2, n_informative=2,
-                           bias=10., random_state=42)
-
-    X_target = arr_type(X)
-
-    ref_model = ridge_regression(X, y, 1,
-                                 solver='cholesky',
-                                 sample_weight=sample_weight,
-                                 return_intercept=return_intercept
-                                 )
+    X = np.random.rand(1000, 3)
+    true_coefs = [1, 2, 0.1]
+    y = np.dot(X, true_coefs)
+    X_testing = arr_type(X)
 
-    tested = ridge_regression(X, y, 1,
+    target = ridge_regression(X_testing, y, 1,
                               solver='auto',
                               sample_weight=sample_weight,
                               return_intercept=return_intercept
                               )
     try:
-        assert_array_almost_equal(ref_model[0], tested[0])
-        assert_almost_equal(ref_model[1], tested[1])
-    except IndexError:
-        assert_array_almost_equal(ref_model[0], tested[0])
-
-
-
+        coef, intercept = target
+        assert_array_almost_equal(coef, true_coefs, decimal=1)
+        assert_array_almost_equal(intercept, 0, decimal=1)
+    except ValueError:
+        assert_array_almost_equal(target, true_coefs, decimal=1)
 
 
 def test_errors_and_values_helper():
