fix: replace deprecated numpy.random.RandomState with Generator API

[sonarqube-agent]

This PR resolves 10 SonarQube S6711 issues in the hierarchical clustering tests by replacing all legacy numpy.random.RandomState calls with the modern numpy.random.default_rng API. This modernization improves code quality and aligns with current NumPy best practices for random number generation.

View Project in SonarCloud

---

Fixed Issues

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:185

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(global_random_seed) call at line 185 with the modern np.random.default_rng(global_random_seed), switching from the deprecated RandomState to the recommended Generator API. This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 185.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -185,1 +185,1 @@ def test_agglomerative_clustering(global_random_seed, lil_container):
-    rng = np.random.RandomState(global_random_seed)
+    rng = np.random.default_rng(global_random_seed)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:277

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(global_random_seed) call at line 277 with the modern np.random.default_rng(global_random_seed). This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 277.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -277,1 +277,1 @@ def test_ward_agglomeration(global_random_seed):
-    rng = np.random.RandomState(global_random_seed)
+    rng = np.random.default_rng(global_random_seed)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:328

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(global_random_seed) call at line 328 with the modern np.random.default_rng(global_random_seed). This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 328.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -328,1 +328,1 @@ def test_sparse_scikit_vs_scipy(global_random_seed):
-    rng = np.random.RandomState(global_random_seed)
+    rng = np.random.default_rng(global_random_seed)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:367

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(global_random_seed) call at line 367 with the modern np.random.default_rng(global_random_seed). This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 367.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -367,1 +367,1 @@ def test_vector_scikit_single_vs_scipy_single(global_random_seed):
-    rng = np.random.RandomState(global_random_seed)
+    rng = np.random.default_rng(global_random_seed)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:466

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(global_random_seed) call at line 466 with the modern np.random.default_rng(global_random_seed). This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 466.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -466,1 +466,1 @@ def test_ward_tree_children_order(global_random_seed):
-    rng = np.random.RandomState(global_random_seed)
+    rng = np.random.default_rng(global_random_seed)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:270

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(0) call at line 270 with np.random.default_rng(0), and updates the associated rng.randn(50, 100) call to rng.standard_normal((50, 100)). This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 270.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -270,2 +270,2 @@ def test_agglomerative_clustering_memory_mapped():
-    rng = np.random.RandomState(0)
-    Xmm = create_memmap_backed_data(rng.randn(50, 100))
+    rng = np.random.default_rng(0)
+    Xmm = create_memmap_backed_data(rng.standard_normal((50, 100)))

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:396

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(seed=1) call at line 396 with the modern np.random.default_rng(seed=1). This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 396.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -396,1 +396,1 @@ def test_mst_linkage_core_memory_mapped(metric_param_grid):
-    rng = np.random.RandomState(seed=1)
+    rng = np.random.default_rng(seed=1)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:157

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(0) call at line 157 with the modern np.random.default_rng(0), switching from the deprecated RandomState to the recommended Generator API. This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 157.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -157,1 +157,1 @@ def test_agglomerative_clustering_distances(
-    rng = np.random.RandomState(0)
+    rng = np.random.default_rng(0)

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:104

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(0) call at line 104 with the modern np.random.default_rng(0), and updates the associated rng.randn(50, 100) call to rng.standard_normal((50, 100)) to use the new Generator API. This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 104.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -104,2 +104,2 @@ def test_unstructured_linkage_tree():
-    rng = np.random.RandomState(0)
-    X = rng.randn(50, 100)
+    rng = np.random.default_rng(0)
+    X = rng.standard_normal((50, 100))

python:S6711 - Use a "numpy.random.Generator" here instead of this legacy function. • MAJOR • View issue

Location: sklearn/cluster/tests/test_hierarchical.py:128

Why is this an issue?

Using a predictable seed is a common best practice when using NumPy to create reproducible results. To that end, using np.random.seed(number) to set the seed of the global numpy.random.RandomState has been the privileged solution for a long time.

What changed

Replaces the legacy np.random.RandomState(0) call at line 128 with the modern np.random.default_rng(0), switching from the deprecated RandomState to the recommended Generator API. This directly fixes the static analysis warning about using legacy numpy.random.RandomState at line 128.

--- a/sklearn/cluster/tests/test_hierarchical.py
+++ b/sklearn/cluster/tests/test_hierarchical.py
@@ -128,1 +128,1 @@ def test_height_linkage_tree():
-    rng = np.random.RandomState(0)
+    rng = np.random.default_rng(0)

Have a suggestion or found an issue? Share your feedback here.

---

SonarQube Remediation Agent uses AI. Check for mistakes.

[sonarqubecloud]

SonarQube reviewer guide

Quality Gate passed

Issues
0 New issues
0 Accepted issues
0 New dependency risks

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud