Commit Graph

8 Commits

Author SHA1 Message Date
NanoCode012
2501c1a6a3 Fix: Gradient Accumulation issue (#1980)
* feat: support new arg num_items_in_batch

* use kwargs to manage extra unknown kwargs for now

* upgrade against upstream transformers main

* make sure trl is on latest too

* fix for upgraded trl

* fix: handle trl and transformer signature change

* feat: update trl to handle transformer signature

* RewardDataCollatorWithPadding no longer has max_length

* handle updated signature for tokenizer vs processor class

* invert logic for tokenizer vs processor class

* processing_class, not processor class

* also handle processing class in dpo

* handle model name w model card creation

* upgrade transformers and add a loss check test

* fix install of tbparse requirements

* make sure to add tbparse to req

* feat: revert kwarg to positional kwarg to be explicit

---------

Co-authored-by: Wing Lian <wing.lian@gmail.com>
2024-10-25 11:28:23 -04:00
Wing Lian
05b398a072 fix some of the edge cases for Jamba (#1452)
* fix some of the edge cases for Jamba

* update requirements for jamba
2024-03-29 02:38:02 -04:00
Mark Saroufim
44ba616da2 Fix broken pypi.yml (#1099) [skip ci] 2024-01-11 12:35:31 -05:00
Wing Lian
772cd870d4 fix the sed command to replace the version w the tag
Some checks failed
pre-commit / pre-commit (push) Has been cancelled
publish pypi / Upload release to PyPI (push) Has been cancelled
PyTest / test (3.10) (push) Has been cancelled
PyTest / test (3.9) (push) Has been cancelled
2023-09-11 13:44:19 -04:00
Wing Lian
bcbc9597e9 replace tags, build dist for pypi publish (#553)
* replace tags, build dist for pypi publish

* missing trailing comma
2023-09-11 13:25:41 -04:00
Wing Lian
20ed4c1f9e pypi on tag push (#552) 2023-09-11 10:33:42 -04:00
Wing Lian
c5dedb17ad remove with section, doesn't seem to work (#551) 2023-09-11 10:27:17 -04:00
Wing Lian
b56503d423 publish to pypi workflow on tagged release (#549) 2023-09-11 09:44:47 -04:00