Closed XinyanLi2016 closed 3 years ago
Thank you for having an interest in our work.
Could you try using PyTorch v1.6.0 and Transformers v2.1.1 instead of the latest versions of them? I believe that this will work.
I'm sorry I'm busy now and I can't immediately fix a bug under the latest versions.
Thanks for your reply, I will try it!
Hi @yahshibu
I have a similar error and I'm using PyTorch v1.6.0 and Transformers v2.1.1
When running train.py
, it runs the first epoch just fine and save a checkpoint in dumps folder, but before executing the second epoch that error appears.
@XinyanLi2016 @mahmoudSalim I updated this repository so that it works with the latest versions, PyTorch (v1.10.0) and Transformers (v4.12.0). I've confirmed that this works on my local computer. Could you update your local code and try it? Sorry for bothering you.
Hi, yahshibu! Recently, I retrain your code with GENIA corpus without fine tuning. But I met an error, which is "IndexError: tuple index out of range". Traceback shows as this:
I can't find why this error happened. Could you help me? Thanks a lot!