scikit-learn · jeremiedbb · Apr 10, 2025 · May 30, 2023 · Jul 1, 2023 · Jul 13, 2023
diff --git a/sklearn/utils/tests/test_optimize.py b/sklearn/utils/tests/test_optimize.py
@@ -3,14 +3,14 @@
 from scipy.optimize import fmin_ncg
 
 from sklearn.exceptions import ConvergenceWarning
-from sklearn.utils._testing import assert_array_almost_equal
+from sklearn.utils._testing import assert_allclose
 from sklearn.utils.optimize import _newton_cg
 
 
-def test_newton_cg():
+def test_newton_cg(global_random_seed):
     # Test that newton_cg gives same result as scipy's fmin_ncg
 
-    rng = np.random.RandomState(0)
+    rng = np.random.RandomState(global_random_seed)
     A = rng.normal(size=(10, 10))
     x0 = np.ones(10)
 
@@ -27,9 +27,13 @@ def hess(x, p):
     def grad_hess(x):
         return grad(x), lambda x: A.T.dot(A.dot(x))
 
-    assert_array_almost_equal(
-        _newton_cg(grad_hess, func, grad, x0, tol=1e-10)[0],
+    # func is a definite positive quadratic form, so the minimum is at x = 0
+    # hence the use of absolute tolerance.
+    assert np.all(np.abs(_newton_cg(grad_hess, func, grad, x0, tol=1e-10)[0]) <= 1e-7)
+    assert_allclose(
+        _newton_cg(grad_hess, func, grad, x0, tol=1e-7)[0],
         fmin_ncg(f=func, x0=x0, fprime=grad, fhess_p=hess),
+        atol=1e-5,
     )