def test_get_all_columns_except(self):
columns = 'ABCD'
to_remove = 'D'
df = create_random_dataframe(rows=4, cols=4, columns=columns)
df = get_all_columns_except(df, to_remove)
self.assertEqual(''.join(list(df)), columns.replace(to_remove, ''))
def test_create_random_dataframe(self):
rows = 5
cols = 5
columns = 'ABCDE'
df = create_random_dataframe(rows, cols, columns)
self.__check_dataframe_data(df, rows, cols, columns)
def test_series_to_list(self):
df = dataframe_utils.create_random_dataframe(rows=5,
cols=5,
columns='ABCDE')
for row in df.iterrows():
output_list = series_to_list(row)
self.assertListEqual(row[1].values.tolist(),
output_list[1].values.tolist())
def test_get_X_y(self):
df = dataframe_utils.create_random_dataframe(rows=5,
cols=5,
columns='ABCDE')
target = 'E'
X_list, y_list = get_X_y(df, target)
# check y values
for index, y_final in enumerate(y_list):
self.assertEqual(
df.loc[:, df.columns == target].values.tolist()[index][0],
y_final)
for index_list, X_list in enumerate(X_list):
for index_item, item_x in enumerate(X_list):
current_df = df.loc[:, df.columns != target].values.tolist(
)[index_list][index_item]
self.assertEqual(current_df, item_x)
def test_make_visualizations_generate_artifacts(self):
labels = [[0], [1]]
data, labels = label_encode(labels)
embeddings = np.array([[1, 0], [0, 1]])
embeddings_list = []
embeddings_list.append(embeddings)
df = dataframe_utils.create_random_dataframe(2, 2, 'AB')
path = 'test_visualization'
make_visualizations(list(labels),
embeddings_list,
df,
output_path=path)
path_created = os.path.exists(path)
self.assertTrue(path_created)
shutil.rmtree(path)
def test_check_target_existent_in_df(self) -> None:
df = dataframe_utils.create_random_dataframe(columns='ABCD')
self.assertRaises(ValueError, check_target_existent_in_df, 'E', df)