Skip to content
This repository has been archived by the owner on Feb 17, 2024. It is now read-only.

mt5-base fine-tuning failed using single A100 GPU #108

Open
pedramyamini opened this issue Sep 1, 2022 · 0 comments
Open

mt5-base fine-tuning failed using single A100 GPU #108

pedramyamini opened this issue Sep 1, 2022 · 0 comments

Comments

@pedramyamini
Copy link

I'm trying to fine-tune mt5-base for summarization task using the script in chapter 7 of huggingface course but it fails at model.fit raising statefulpartitionedcall error using Nvidia A100 40 GB VRAM but works well for mt5-small. What is the problem? Any solution? Thank you.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant