How to use the rnnmorph.settings.TEST_UNTAGGED_JZ function in rnnmorph

def tag_ru_files(predictor: RNNMorphPredictor) -> Dict:
    if not os.path.exists(TEST_TAGGED_FOLDER):
        os.makedirs(TEST_TAGGED_FOLDER)
    tag(predictor, TEST_UNTAGGED_LENTA, TEST_TAGGED_LENTA)
    tag(predictor, TEST_UNTAGGED_VK, TEST_TAGGED_VK)
    tag(predictor, TEST_UNTAGGED_JZ, TEST_TAGGED_JZ)

    quality = dict()
    print("Lenta:")
    quality['Lenta'] = measure(TEST_GOLD_LENTA, TEST_TAGGED_LENTA, True, None)
    print("VK:")
    quality['VK'] = measure(TEST_GOLD_VK, TEST_TAGGED_VK, True, None)
    print("JZ:")
    quality['JZ'] = measure(TEST_GOLD_JZ, TEST_TAGGED_JZ, True, None)
    print("All:")
    count_correct_tags = quality['Lenta'].correct_tags + quality['VK'].correct_tags + quality['JZ'].correct_tags
    count_correct_pos = quality['Lenta'].correct_pos + quality['VK'].correct_pos + quality['JZ'].correct_pos
    count_tags = quality['Lenta'].total_tags + quality['VK'].total_tags + quality['JZ'].total_tags
    count_correct_sentences = quality['Lenta'].correct_sentences + quality['VK'].correct_sentences + \
                              quality['JZ'].correct_sentences
    count_sentences = quality['Lenta'].total_sentences + quality['VK'].total_sentences + \
                      quality['JZ'].total_sentences