You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 17, 2024. It is now read-only.
I am using mt5-small for machine translation task using pytorch and transformer. I have approx 3 million parallel data and using 96GB RAM and 1 P-100 GPU for fine-tuning, but not able to fine tuning because RAM are fully exhausted before training started.
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I am using mt5-small for machine translation task using pytorch and transformer. I have approx 3 million parallel data and using 96GB RAM and 1 P-100 GPU for fine-tuning, but not able to fine tuning because RAM are fully exhausted before training started.
The text was updated successfully, but these errors were encountered: