COSMIT use explicit RandomState in KDE tests

Author: larsmans

sklearn/neighbors/tests/test_kde.py

@@ -26,9 +26,9 @@ def compute_kernel_slow(Y, X, kernel, h):
 
 
 def test_kernel_density(n_samples=100, n_features=3):
-    np.random.seed(0)
-    X = np.random.random((n_samples, n_features))
-    Y = np.random.random((n_samples, n_features))
+    rng = np.random.RandomState(0)
+    X = rng.randn(n_samples, n_features)
+    Y = rng.randn(n_samples, n_features)
 
     for kernel in ['gaussian', 'tophat', 'epanechnikov',
                    'exponential', 'linear', 'cosine']:
@@ -52,8 +52,8 @@ def check_results(kernel, bandwidth, atol, rtol):
 
 
 def test_kernel_density_sampling(n_samples=100, n_features=3):
-    np.random.seed(0)
-    X = np.random.random((n_samples, n_features))
+    rng = np.random.RandomState(0)
+    X = rng.randn(n_samples, n_features)
 
     bandwidth = 0.2
 
@@ -80,16 +80,16 @@ def test_kernel_density_sampling(n_samples=100, n_features=3):
         assert_raises(NotImplementedError, kde.sample, 100)
 
     # non-regression test: used to return a scalar
-    X = np.random.random((4, 1))
+    X = rng.randn(4, 1)
     kde = KernelDensity(kernel="gaussian").fit(X)
     assert_equal(kde.sample().shape, (1, 1))
 
 
 def test_kde_algorithm_metric_choice():
     """Smoke test for various metrics and algorithms"""
-    np.random.seed(0)
-    X = np.random.random((10, 2))  # 2 features required for haversine dist.
-    Y = np.random.random((10, 2))
+    rng = np.random.RandomState(0)
+    X = rng.randn(10, 2)    # 2 features required for haversine dist.
+    Y = rng.randn(10, 2)
 
     for algorithm in ['auto', 'ball_tree', 'kd_tree']:
         for metric in ['euclidean', 'minkowski', 'manhattan',