def test_predict(self):
clf = KMeans(epsilon=1, bounds=(0, 1), n_clusters=3)
X = np.array([0.1, 0.1, 0.1, 0.1, 0.5, 0.5, 0.5, 0.5, 0.9, 0.9, 0.9]).reshape(-1, 1)
clf.fit(X)
predicted = clf.predict(np.array([0.1, 0.5, 0.9]).reshape(-1, 1))
# print("Predictions: %s" % str(predicted))
# print("Centers: %s" % str(clf.cluster_centers_))
self.assertTrue(0 <= predicted[0] <= 2)
self.assertTrue(0 <= predicted[1] <= 2)
self.assertTrue(0 <= predicted[2] <= 2)
def test_many_features(self):
X = np.random.random(size=(500, 3))
bounds = (0, 1)
clf = KMeans(bounds=bounds, n_clusters=4)
clf.fit(X)
centers = clf.cluster_centers_
self.assertEqual(centers.shape[0], 4)
self.assertEqual(centers.shape[1], 3)
self.assertTrue(np.all(clf.transform(X).argmin(axis=1) == clf.predict(X)))
def test_predict(self):
clf = KMeans(30, (0, 1), 3)
X = np.array([0.1, 0.1, 0.1, 0.1, 0.5, 0.5, 0.5, 0.5, 0.9, 0.9,
0.9]).reshape(-1, 1)
clf.fit(X)
predicted = clf.predict(np.array([0.1, 0.5, 0.9]).reshape(-1, 1))
# print("Predictions: %s" % str(predicted))
# print("Centers: %s" % str(clf.cluster_centers_))
self.assertNotEqual(predicted[0], predicted[1])
self.assertNotEqual(predicted[0], predicted[2])
self.assertNotEqual(predicted[2], predicted[1])