在ai studio上进行自定义数据的SQUAD型阅读理解,报错Exception: the train file: train.json only has one column but it is not a predict file,请问怎么改
Created by: wangdanana
[2020-03-23 11:24:54,689] [ INFO] - Installing bert_chinese_L-12_H-768_A-12 module [2020-03-23 11:24:54,693] [ INFO] - Module bert_chinese_L-12_H-768_A-12 already installed in /home/aistudio/.paddlehub/modules/bert_chinese_L_12_H_768_A_12 [2020-03-23 11:24:58,420] [ INFO] - Load pretraining parameters from /home/aistudio/.paddlehub/modules/bert_chinese_L_12_H_768_A_12/assets/params. Traceback (most recent call last): File "reading_comprehension.py", line 122, in dataset = DemoDataset() File "reading_comprehension.py", line 88, in init train_file="train.json", File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddlehub/dataset/base_nlp_dataset.py", line 51, in init predict_file_with_header=predict_file_with_header) File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddlehub/dataset/dataset.py", line 94, in init self._load_train_examples() File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddlehub/dataset/dataset.py", line 150, in _load_train_examples self.train_examples = self._read_file(self.train_path, phase="train") File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddlehub/dataset/base_nlp_dataset.py", line 68, in _read_file % (phase, input_file)) Exception: the train file: train.json only has one column but it is not a predict file