Error creating instruct_model: NameError: name 'AutoModelForSeq2SeqLM' is not defined

Course Q&A Generative AI with Large Language Models
1

I am getting this error in Lab 2.2:

NameError Traceback (most recent call last)
in
----> 1 instruct_model = AutoModelForSeq2SeqLM.from_pretrained(“./flan-dialogue-summary-checkpoint”, torch_dtype=torch.bfloat16)

NameError: name ‘AutoModelForSeq2SeqLM’ is not defined

This prevents me from proceeding. Any ideas?
Thanks!

2

Yeah, have you run the cell where it is defined (search above) or have you imported the all the imports?

3

Yes but my kernel failed. I had to restart with the proper 8vCPU + 32Gi and rerun the entire lab. now it works.

1 Like