Skip to content

Conversation

@kmehant
Copy link
Collaborator

@kmehant kmehant commented Sep 24, 2025

Design

Screenshot 2025-09-25 at 4 52 58 AM

Unit tests

Unit tests are added to test

  1. reward computation correctness
  2. dataset's ability to learn from rewards.

E2E functional tests are added in the fms-hf-tuning repository PR which includes mixing 2 fine-tuning datasets across multiple settings.

Summary of unit tests in fms-acceleration
Screenshot 2025-09-25 at 4 00 10 AM

Summary of test in fms-hf-tuning PR
Screenshot 2025-09-25 at 10 30 38 PM

Screenshot 2025-09-25 at 10 30 16 PM

Future TODOs

see issue #153

Co-authored-by: romit <[email protected]>
Co-authored-by: Padmanabha V Seshadri <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Co-authored-by: romit <[email protected]>
Co-authored-by: Padmanabha V Seshadri <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
@kmehant kmehant changed the title feat: add online data mixing plugin [DO NOT MERGE] feat: add online data mixing plugin Sep 24, 2025
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
Signed-off-by: Mehant Kammakomati <[email protected]>
@kmehant kmehant changed the title [DO NOT MERGE] feat: add online data mixing plugin feat: add online data mixing plugin Sep 25, 2025
Signed-off-by: Mehant Kammakomati <[email protected]>
@ashokponkumar ashokponkumar merged commit fbf12cb into foundation-model-stack:main Sep 26, 2025
8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants