def test_perform_clustering(self):
test_dir_name = os.path.dirname(__file__)
feat_array_fn = os.path.join(test_dir_name, "data",
"feature_array.csv")
tmp_dir = self.__class__.TMP_DIR
output_fn = os.path.join(tmp_dir, "code.txt")
clusterer = Clusterer(feat_array_fn, output_fn, [])
clusterer.iter = 100
clusterer.perform_clustering()
sample_output_filename = os.path.join(test_dir_name, "data",
"sample_cluster_code.txt")
sample_inv_output_filename = os.path.join(
test_dir_name, "data", "sample_cluster_code_inv.txt")
same_as_sample = filecmp.cmp(output_fn, sample_output_filename)
same_as_inv = filecmp.cmp(output_fn, sample_inv_output_filename)
self.assertTrue(same_as_sample or same_as_inv)
# Now test the data set with non feat cols
feat_array_fn = os.path.join(test_dir_name, "data",
"feature_array_with_non_feat_cols.csv")
clusterer = Clusterer(feat_array_fn, output_fn, ["id", "param"])
clusterer.iter = 100
clusterer.perform_clustering()
same_as_sample = filecmp.cmp(output_fn, sample_output_filename)
same_as_inv = filecmp.cmp(output_fn, sample_inv_output_filename)
self.assertTrue(same_as_sample or same_as_inv)
def test_perform_clustering(self):
test_dir_name = os.path.dirname(__file__)
feat_array_fn = os.path.join(
test_dir_name, "data", "feature_array.csv")
tmp_dir = self.__class__.TMP_DIR
output_fn = os.path.join(tmp_dir, "code.txt")
clusterer = Clusterer(feat_array_fn, output_fn, [])
clusterer.iter = 100
clusterer.perform_clustering()
sample_output_filename = os.path.join(
test_dir_name, "data", "sample_cluster_code.txt")
sample_inv_output_filename = os.path.join(
test_dir_name, "data", "sample_cluster_code_inv.txt")
same_as_sample = filecmp.cmp(output_fn, sample_output_filename)
same_as_inv = filecmp.cmp(output_fn, sample_inv_output_filename)
self.assertTrue(same_as_sample or same_as_inv)
# Now test the data set with non feat cols
feat_array_fn = os.path.join(
test_dir_name, "data", "feature_array_with_non_feat_cols.csv")
clusterer = Clusterer(feat_array_fn, output_fn, ["id", "param"])
clusterer.iter = 100
clusterer.perform_clustering()
same_as_sample = filecmp.cmp(output_fn, sample_output_filename)
same_as_inv = filecmp.cmp(output_fn, sample_inv_output_filename)
self.assertTrue(same_as_sample or same_as_inv)
