运行main.py --model bert_bilstm_crf_adv --data msra,msr 时报错：

[dsxiangli]

运行main.py --model bert_bilstm_crf_adv --data msra,msr 时报错：

[desfc]

报错内容：
………………
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[0,0] = 57 is not in [0, 49)
[[{{node task2_msr/crf_layer/GatherV2_1}}]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "E:/kg_projects/ChineseNER-bert-bilstm-crf/ChineseNER-main/main.py", line 150, in
multitask_train(args)
File "E:/kg_projects/ChineseNER-bert-bilstm-crf/ChineseNER-main/main.py", line 110, in multitask_train
tf.estimator.train_and_evaluate(estimator, train_spec, eval_spec)
File "C:\Users\return0.conda\envs\ChineseNER-main\lib\site-packages\tensorflow_estimator\python\estimator\training.py", line 473, in train_and_evaluate
return executor.run()
………………
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[0,0] = 57 is not in [0, 49)
[[node task2_msr/crf_layer/GatherV2_1 (defined at E:\kg_projects\ChineseNER-bert-bilstm-crf\ChineseNER-main\tools\layer.py:126) ]]
………………

请问是什么原因呀？

[desfc]

感觉出现这个报错的原因是msr数据集。因为多任务NER只要使用msr数据集就会报上面的错，但是单任务NER使用msr数据集却不会报错。
此外使用bert_bilstm_crf_adv模型进行其它多任务NER并不会报这个错，msra数据集用在其它NER任务中也不会报错

[DSXiangLi]

@fengxuefx 辛苦发一下你的运行脚本吧