ros/LlamaFactory
1
0
Fork 0
mirror of https://github.com/hiyouga/LlamaFactory.git synced 2026-03-20 11:43:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

alter rewards data type

Browse Source 
Former-commit-id: 3eb7eb2d37525da50fe401ab7c59532e6e1ef984
This commit is contained in:
hiyouga
2023-06-02 14:19:51 +08:00
parent 896dbfec16
commit e9ab06678f
12 changed files with 40 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/train_ppo.py
View File

@@ -70,7 +70,7 @@ def main():
ppo_trainer.save_model()
ppo_trainer.save_state() # must be after save_model
if ppo_trainer.is_world_process_zero() and model_args.plot_loss:
plot_loss(training_args, keys=["loss", "reward"])
plot_loss(training_args.output_dir, keys=["loss", "reward"])
def _mp_fn(index):