RuntimeError: The size of the 2D attn_mask is not correct.

[YaxinCountry]

Hi urchade,

Thanks a lot for this great work.
I'm trying to reproduce your results on CONLL04. But I found that the code reported an error:
Arguments: ("RuntimeError: The size of the 2D attn_mask is not correct.")
Through debugging, The error is reported in the "loss = model(batch)"l in the train.py file.In the source code, this statement is placed in a try-catch, which will cause the model to not be trained, how can I fix this?

Looking forward to your reply, thank you.

Sincerely

[urchade]

Hi,

I think this is due you use of different library version as mine. I have used flair==0.11.3, torch=1.10.1 and transformer==4.29.1

Urchade

[M0Ta0]

Hi urchade,

When I installed allennlp, it was not compatible with the Transformers and Flair versions.
How to solve this problem？

Looking forward to your reply, thank you.Wishing you a better future
sincerely