thomasjpfan
diff --git a/‎sklearn/linear_model/_ridge.py
Lines changed: 4 additions & 2 deletions b/‎sklearn/linear_model/_ridge.py
Lines changed: 4 additions & 2 deletions
diff --git a/‎sklearn/linear_model/tests/test_coordinate_descent.py
Lines changed: 5 additions & 2 deletions b/‎sklearn/linear_model/tests/test_coordinate_descent.py
Lines changed: 5 additions & 2 deletions
diff --git a/‎sklearn/linear_model/tests/test_ridge.py
Lines changed: 21 additions & 0 deletions b/‎sklearn/linear_model/tests/test_ridge.py
Lines changed: 21 additions & 0 deletions
@@ -1483,7 +1483,7 @@ def __init__(
         is_clf=False,
         alpha_per_target=False,
     ):
-        self.alphas = np.asarray(alphas)
+        self.alphas = alphas
         self.fit_intercept = fit_intercept
         self.normalize = normalize
         self.scoring = scoring
@@ -1842,6 +1842,8 @@ def fit(self, X, y, sample_weight=None):
         if sample_weight is not None:
             sample_weight = _check_sample_weight(sample_weight, X, dtype=X.dtype)
 
+        self.alphas = np.asarray(self.alphas)
+
         if np.any(self.alphas <= 0):
             raise ValueError(
                 "alphas must be strictly positive. Got {} containing some "
@@ -1977,7 +1979,7 @@ def __init__(
         store_cv_values=False,
         alpha_per_target=False,
     ):
-        self.alphas = np.asarray(alphas)
+        self.alphas = alphas
         self.fit_intercept = fit_intercept
         self.normalize = normalize
         self.scoring = scoring
 
@@ -351,7 +351,7 @@ def test_lasso_cv_positive_constraint():
 
 def _scale_alpha_inplace(estimator, n_samples):
     """Rescale the parameter alpha from when the estimator is evoked with
-    normalize set to True to when it is evoked in a Pipeline with normalize set
+    normalize set to True as if it were evoked in a Pipeline with normalize set
     to False and with a StandardScaler.
     """
     if ("alpha" not in estimator.get_params()) and (
@@ -360,7 +360,10 @@ def _scale_alpha_inplace(estimator, n_samples):
         return
 
     if isinstance(estimator, (RidgeCV, RidgeClassifierCV)):
-        alphas = estimator.alphas * n_samples
+        # alphas is not validated at this point and can be a list.
+        # We convert it to a np.ndarray to make sure broadcasting
+        # is used.
+        alphas = np.asarray(estimator.alphas) * n_samples
         return estimator.set_params(alphas=alphas)
     if isinstance(estimator, (Lasso, LassoLars, MultiTaskLasso)):
         alpha = estimator.alpha * np.sqrt(n_samples)
 
@@ -1110,6 +1110,27 @@ def test_ridge_classifier_cv_store_cv_values(scoring):
     assert r.cv_values_.shape == (n_samples, n_targets, n_alphas)
 
 
+@pytest.mark.parametrize("Estimator", [RidgeCV, RidgeClassifierCV])
+def test_ridgecv_alphas_conversion(Estimator):
+    rng = np.random.RandomState(0)
+    alphas = (0.1, 1.0, 10.0)
+
+    n_samples, n_features = 5, 5
+    if Estimator is RidgeCV:
+        y = rng.randn(n_samples)
+    else:
+        y = rng.randint(0, 2, n_samples)
+    X = rng.randn(n_samples, n_features)
+
+    ridge_est = Estimator(alphas=alphas)
+    assert (
+        ridge_est.alphas is alphas
+    ), f"`alphas` was mutated in `{Estimator.__name__}.__init__`"
+
+    ridge_est.fit(X, y)
+    assert_array_equal(ridge_est.alphas, np.asarray(alphas))
+
+
 def test_ridgecv_sample_weight():
     rng = np.random.RandomState(0)
     alphas = (0.1, 1.0, 10.0)