def test_determine_k(self):
"""
Test the clusterer._dtermine_k function.
"""
test_dir_name = os.path.dirname(__file__)
feat_array_fn = os.path.join(test_dir_name, "data",
"four_clusters.csv")
df = pd.read_csv(feat_array_fn)
feat_array = df[["x", "y"]].values
clusterer = Clusterer(feat_array_fn, "/dev/null", [])
best_k = clusterer._determine_k(feat_array, 9)
self.assertEqual(best_k, 4)
feat_array_fn = os.path.join(test_dir_name, "data", "iris.csv")
df = pd.read_csv(feat_array_fn)
feat_array = df[[
"Sepal.Length", "Sepal.Width", "Petal.Length", "Petal.Width"
]].values
clusterer = Clusterer(feat_array_fn, "/dev/null", [])
best_k = clusterer._determine_k(feat_array, 9)
self.assertEqual(best_k, 2)
def test_determine_k(self):
"""
Test the clusterer._dtermine_k function.
"""
test_dir_name = os.path.dirname(__file__)
feat_array_fn = os.path.join(
test_dir_name, "data", "four_clusters.csv")
df = pd.read_csv(feat_array_fn)
feat_array = df[["x", "y"]].values
clusterer = Clusterer(feat_array_fn, "/dev/null", [])
best_k = clusterer._determine_k(feat_array, 9)
self.assertEqual(best_k, 4)
feat_array_fn = os.path.join(
test_dir_name, "data", "iris.csv")
df = pd.read_csv(feat_array_fn)
feat_array = df[[
"Sepal.Length", "Sepal.Width", "Petal.Length",
"Petal.Width"]].values
clusterer = Clusterer(feat_array_fn, "/dev/null", [])
best_k = clusterer._determine_k(feat_array, 9)
self.assertEqual(best_k, 2)
