Satya Almasian

Results 7 comments of Satya Almasian

There is no evaluation available for this model. The model is trained on section structure of wikipedia dump, there is no train and test split.

This happens when the output is a list and not a single value, change the last line of openie.py to : `try: e=json.loads(r.text) except: e=r.text return e` and then it...

I can access the model. Can you tell me what is the issue ?

here is the updated script: https://github.com/satya77/Transformer_Temporal_Tagger/blob/master/examples/train_token_classifier.py

How did you solve it? I have the same issue with LoRA, with or without gradient checkpointing.

What is the format that the data should look like?