423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-29 18:20:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

[trainer] fix pt loss (#7748)

Browse Source 
* fix pt loss

* robust

* fix

* test
This commit is contained in:
hoshi-hiyouga
2025-04-17 03:15:35 +08:00
committed by GitHub
parent a0818eae58
commit 8208cbf1dc
10 changed files with 34 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llamafactory/train/sft/trainer.py
View File

@@ -60,6 +60,8 @@ class CustomSeq2SeqTrainer(Seq2SeqTrainer):
super().__init__(**kwargs)
if processor is not None:
# avoid wrong loss under gradient accumulation
# https://github.com/huggingface/transformers/pull/36044#issuecomment-2746657112
self.model_accepts_loss_kwargs = False
self.finetuning_args = finetuning_args