while running run_qa.py, ran into a value error

command:

python3 run_qa.py --model_name_or_path hyunwoongko/kobart --dataset_name squad_kor_v2 --do_train --do_eval --per_device_train_batch_size 8 --learning_rate 3e-5 --num_train_epochs 3 --max_seq_length 512 --doc_stride 128 --output_dir /tmp/debug_squad/

error:

ValueError: External features info don't match the dataset:
Got
{'id': Value(dtype='string', id=None), 'title': Value(dtype='string', id=None), 'context': Value(dtype='string', id=None), 'question': Value(dtype='string', id=None), 'answer': {'text': Value(dtype='string', id=None), 'answer_start': Value(dtype='int32', id=None), 'html_answer_start': Value(dtype='int32', id=None)}, 'url': Value(dtype='string', id=None), 'raw_html': Value(dtype='string', id=None)}
with type
struct, context: string, id: string, question: string, raw_html: string, title: string, url: string>

but expected something like
{'answer': {'answer_start': Value(dtype='int32', id=None), 'html_answer_start': Value(dtype='int32', id=None), 'text': Value(dtype='string', id=None)}, 'context': Value(dtype='string', id=None), 'id': Value(dtype='string', id=None), 'question': Value(dtype='string', id=None), 'raw_html': Value(dtype='string', id=None), 'title': Value(dtype='string', id=None), 'url': Value(dtype='string', id=None)}
with type
struct, context: string, id: string, question: string, raw_html: string, title: string, url: string>

I didn't encounter this error 4 hours ago. any solutions for this kind of issue?
looks like gained dataset format refers to 'Data Fields', while expected refers to 'Data Instances'.