데이터셋 구축 과정에서 에러가 발생합니다
Closed this issue · 0 comments
안녕하세요, CD_dataset.py 구동하는데 에러가 발생해서 문의드립니다.
Traceback (most recent call last):
File "/absa/Korean_ABSA_model/scripts/CD_pipeline.py", line 42, in
dataset_train, dataset_dev, dataset_test = get_CD_dataset(train_data, dev_data, test_data,
File "/absa/Korean_ABSA_model/scripts/CD_dataset.py", line 175, in get_CD_dataset
train_CD_data, train_SC_data = CD_dataset(train_data, tokenizer, max_len)
File "/absa/Korean_ABSA_model/scripts/CD_dataset.py", line 54, in CD_dataset
entity_property_data_dict, polarity_data_dict = tokenize_and_align_labels(tokenizer, form, utterance['annotation'], max_len)
File "/absa/Korean_ABSA_model/scripts/CD_dataset.py", line 146, in tokenize_and_align_labels
first_sep = tokenized_data['input_ids'].index(3)
ValueError: 3 is not in list