Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[FEATURE] add CarperAI/trlx promp-completion support to ArgillaTrainer #3378

Closed
davidberenstein1957 opened this issue Jul 11, 2023 · 1 comment
Assignees
Labels
area: trainer Indicates that an issue or pull request is related to the Argilla Trainer type: enhancement Indicates new feature requests

Comments

@davidberenstein1957
Copy link
Member

davidberenstein1957 commented Jul 11, 2023

Is your feature request related to a problem? Please describe.
I cannot do RewardModelling with the TextField or TextQuestion.

Describe the solution you'd like
We should add support for https://github.com/CarperAI/trlx.
This support should include

  • a .for_supervised_fine_tuning() class method for the TrainingTaskMapping.
  • a method to support TextField and TextQuestion function for Promp-Completsion
  • support for the prepare_for_training() method of the FeedbackDataset
  • support for the FrameWork(Enum) via trlx
  • alignment with our docs

Describe alternatives you've considered
consider taking lvwerra/trl into account too. #3377

Additional context
N.A.

@davidberenstein1957 davidberenstein1957 added the type: enhancement Indicates new feature requests label Jul 11, 2023
@tomaarsen tomaarsen self-assigned this Jul 17, 2023
@davidberenstein1957 davidberenstein1957 added the area: trainer Indicates that an issue or pull request is related to the Argilla Trainer label Aug 28, 2023
@tomaarsen
Copy link
Contributor

Lower usage and value than TRL, so we'll disregarding this for now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area: trainer Indicates that an issue or pull request is related to the Argilla Trainer type: enhancement Indicates new feature requests
Projects
None yet
Development

No branches or pull requests

2 participants