Mujeen Sung

Thanks for reporting the error. Could you replace line 13--16 as follows? ``` entity_id = str(i) entity_name = f[entity_id].attrs['text'] embedding = f[entity_id]['embedding'][:] ```

Hi, @seasonyao Thanks for reaching out to us. Basically, Bio-LM described in our paper uses RoBERTa as a backbone language model, and we observed that Bio-LM outperforms BioBERT on BioLAMA....

Hi @Kik099 The instructions for training NER models are described at https://github.com/dmis-lab/BERN2/tree/main/multi_ner/training. You can review the format of the training data there. If you encounter any issues while training models...

Sorry for the inconvenience. It seems that the timeout issue occasionally arises due to heavy server traffic. I would like to suggest considering a local installation of BERN2.

Hi @jessicapatricoski Sorry for the inconvenience. I suggest upgrading your PyTorch version to ensure compatibility with your GPU. Could you please let us know if BERN2 works after upgrading PyTorch?

Hi @majdabd Could you please provide any abstracts where you have encountered the issue? Also, does the BERN2 web demo annotate mutation entities when you input the abstract?

Thanks for sharing! I'll try reproducing and figuring out the issue from my side and get back to you.

Thanks @DunCer for sharing your solution!