[WIP] Add Minimum Risk Trainer support#427

Draft

alexandremuzio wants to merge 9 commits intoCarperAI:mainfrom alexandremuzio:mrt

+1,155-2

Commits on Mar 7, 2023

[WIP] MRT
alexandremuzio
committed

Commits on Mar 15, 2023

[WIP] make_experience working
Alex Muzio
committed

Commits on Mar 16, 2023

[WIP] MRT make_experience working with batch
Alex Muzio
committed

Commits on Mar 17, 2023

[WIP] MRT Training now actually training
Alex Muzio
committed

Commits on Mar 21, 2023

Creating debug file separately for MRT T5
Alex Muzio
committed

Commits on Mar 22, 2023

Adding T5 translation task with ppo
Alex Muzio
committed

Commits on Mar 23, 2023

Adding metric_fn to ppo translation example
Alex Muzio
committed

Commits on Apr 9, 2023

Merge branch 'main' into mrt
alexandremuzio
committed

Commits on Apr 10, 2023

Removing some unused stuff + some improvements + fixing formatting
alexandremuzio
committed