[WIP] Add Minimum Risk Trainer support #427
Draft
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Initial working version of MRT.
Besides the difference in the learning procedure, most of the MRT types are pretty similar to the PPO ones, such as
MRTRLElement. The main difference is that MRT generates a batch of candidate sentences for the same prompt instead of a single sentence per prompt.Main changes:
trlx/data/mrt_types.py,examples/mrt_translation_t5.pyI've currently tested it with t5 on translation and summarization tasks.
TODOs:
MarianMTmodelsdebug_mrt_summarize_daily_cnn_t5.py(only for debugging)Please let me know if there are any other suggestions as well. cc @LouisCastricato