tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

chore: refactor if condition

Browse Source
This commit is contained in:
NanoCode012
2025-05-23 12:47:13 +07:00
parent 36234dc746
commit 29dbf98ed2
1 changed files with 3 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/axolotl/core/trainer_builder/rl.py
View File

@@ -183,9 +183,8 @@ class HFRLTrainerBuilder(TrainerBuilderBase):
trainer_kwargs["label_smoothing"] = self.cfg.dpo_label_smoothing
if self.eval_dataset:
trainer_kwargs["eval_dataset"] = self.eval_dataset
if self.cfg.adapter and self.peft_config:
if self.cfg.rl is not RLType.GRPO:
trainer_kwargs["peft_config"] = self.peft_config
if self.cfg.adapter and self.peft_config and self.cfg.rl is not RLType.GRPO:
trainer_kwargs["peft_config"] = self.peft_config
if self.cfg.precompute_ref_log_probs is not None:
trainer_kwargs["precompute_ref_log_probs"] = (
self.cfg.precompute_ref_log_probs
@@ -243,5 +242,5 @@ class HFPPOTrainerBuilder(TrainerBuilderBase):
return callbacks
def build(self, total_num_steps):
# build PPOConfig
# TODO: build PPOConfig
raise NotImplementedError("PPO trainer builder is not implemented yet.")