解决“TypeError: prepare_inputs_for_inference() got an unexpected keyword argument ‘cache_position‘”问题

E:\Anaconda\envs\BERTCHINESE\python.exe C:\Users\asus\Desktop\毕设\bert-base-chinese\1.2.py Some weights of DualTaskBert were not initialized from the model checkpoint at C:\Users\asus\Desktop\毕设\bert-base-chinese and are newly initialized: ['classifier.1.bias', 'classifier.1.weight', 'regressor.bias', 'regressor.weight'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. Epoch 1 | Loss: 7838.2666 Epoch 2 | Loss: 7616.3560 Epoch 3 | Loss: 7447.6118 Epoch 4 | Loss: 7218.7124 Traceback (most recent call last): File "C:\Users\asus\Desktop\毕设\bert-base-chinese\1.2.py", line 138, in <module> print(predict(test_title, test_abstract)) File "C:\Users\asus\Desktop\毕设\bert-base-chinese\1.2.py", line 128, in predict score, innov_logits = model(**encoding) File "E:\Anaconda\envs\BERTCHINESE\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "E:\Anaconda\envs\BERTCHINESE\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl return forward_call(*args, **kwargs) TypeError: forward() got an unexpected keyword argument 'token_type_ids' Epoch 5 | Loss: 7045.9922 进程已结束，退出代码为 1 这个问题怎么解决

如何解决TypeError: __init__() got an unexpected keyword argument 'service'