CPO / DPO: Fix red CI - Githubissues

huggingface / trl

Train transformer language models with reinforcement learning.

http://hf.co/docs/trl

Apache License 2.0

8.61k stars 1.06k forks source link

CPO / DPO: Fix red CI #1749

Closed younesbelkada closed 1 week ago

younesbelkada commented 1 week ago

I accidentally pushed the fix of the red CI on main. https://github.com/huggingface/trl/pull/1734 overrides TrainingArguments __post_init__ method without returning super().__post_init__().

https://github.com/huggingface/trl/commit/4e23d958f20fd4fdd795cb06c2cdb7ebea704855 combined with this PR should make the CI happy cc @kashif @vwxyzjn

vwxyzjn commented 1 week ago

LGTM! Thanks @younesbelkada!

HuggingFaceDocBuilderDev commented 1 week ago

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.