Use pytest

src/_test/test_umap.py

@@ -1,45 +1,39 @@
 import numpy as np
-import unittest
+import pytest
 
 from mlex_dimension_reduction_pca.umap_run import computeUMAP
 
 
-class TestComputeUMAP(unittest.TestCase):
-
-    def test_1d_input(self):
-        data = np.random.rand(100, 50)
-        result = computeUMAP(data, n_components=2)
-        self.assertEqual(result.shape, (100, 2), "Output shape should be (N, 2)")
-
-    def test_2d_input(self):
-        data = np.random.rand(100, 10, 10)
-        result = computeUMAP(data, n_components=2)
-        self.assertEqual(result.shape, (100, 2), "Output shape should be (N, 2)")
-
-    def test_n_components_3(self):
-        data = np.random.rand(100, 50)
-        result = computeUMAP(data, n_components=3)
-        self.assertEqual(result.shape, (100, 3), "Output shape should be (N, 3) when n_components=3")
-
-    def test_min_dist(self):
-        data = np.random.rand(100, 50)
-        result1 = computeUMAP(data, min_dist=0.1)
-        result2 = computeUMAP(data, min_dist=0.5)
-        self.assertNotEqual(np.sum(result1), np.sum(result2), "UMAP results should differ for different min_dist values")
-
-    def test_random_state(self):
-        data = np.random.rand(100, 50)
-        result1 = computeUMAP(data, random_state=42)
-        result2 = computeUMAP(data, random_state=42)
-        np.testing.assert_array_almost_equal(result1, result2, err_msg="Results should be the same for the same random_state")
-
-    def test_different_random_state(self):
-        data = np.random.rand(100, 50)
-        result1 = computeUMAP(data, random_state=42)
-        result2 = computeUMAP(data, random_state=24)
-        with self.assertRaises(AssertionError):
-            np.testing.assert_array_almost_equal(result1, result2, err_msg="Results should differ for different random_states")
-
-
-if __name__ == '__main__':
-    unittest.main()
+def test_1d_input():
+    data = np.random.rand(100, 50)
+    result = computeUMAP(data, n_components=2)
+    assert result.shape == (100, 2), "Output shape should be (N, 2)"
+
+def test_2d_input():
+    data = np.random.rand(100, 10, 10)
+    result = computeUMAP(data, n_components=2)
+    assert result.shape == (100, 2), "Output shape should be (N, 2)"
+
+def test_n_components_3():
+    data = np.random.rand(100, 50)
+    result = computeUMAP(data, n_components=3)
+    assert result.shape == (100, 3), "Output shape should be (N, 3) when n_components=3"
+
+def test_min_dist():
+    data = np.random.rand(100, 50)
+    result1 = computeUMAP(data, min_dist=0.1)
+    result2 = computeUMAP(data, min_dist=0.5)
+    assert np.sum(result1) != np.sum(result2), "UMAP results should differ for different min_dist values"
+
+def test_random_state():
+    data = np.random.rand(100, 50)
+    result1 = computeUMAP(data, random_state=42)
+    result2 = computeUMAP(data, random_state=42)
+    np.testing.assert_array_almost_equal(result1, result2, err_msg="Results should be the same for the same random_state")
+
+def test_different_random_state():
+    data = np.random.rand(100, 50)
+    result1 = computeUMAP(data, random_state=42)
+    result2 = computeUMAP(data, random_state=24)
+    with pytest.raises(AssertionError):
+        np.testing.assert_array_almost_equal(result1, result2, err_msg="Results should differ for different random_states")