ros/LLaMA-Factory
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #2525 from stephen-nju/main

Browse Source 
update project_kwargs for ppo config

Former-commit-id: e7a6910141cc8d8dd966c1f54388d9ef764418d0
This commit is contained in:
hoshi-hiyouga
2024-02-25 15:54:00 +08:00
committed by GitHub
parent a85bdcf2f6 823f618cba
commit 88f3358320
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
src/llmtuner/train/ppo/workflow.py
View File

@@ -61,6 +61,7 @@ def run_ppo(
use_score_norm=finetuning_args.ppo_score_norm, use_score_norm=finetuning_args.ppo_score_norm,
whiten_rewards=finetuning_args.ppo_whiten_rewards, whiten_rewards=finetuning_args.ppo_whiten_rewards,
accelerator_kwargs={"step_scheduler_with_optimizer": False}, accelerator_kwargs={"step_scheduler_with_optimizer": False},
project_kwargs={"logging_dir": training_args.logging_dir},
) )
# Create optimizer and scheduler # Create optimizer and scheduler