labels = df[1]
train_features, test_features, train_labels, test_labels = train_test_split(features, labels)