423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-10-16 00:28:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix tokenizer padding side in evaluate.py

Browse Source 
Former-commit-id: bcb43ff8ba1946c1f7e7865c9d0fb47ba276935d
This commit is contained in:
hiyouga 2023-10-21 00:30:04 +08:00
parent d602f06882
commit c0658711ca
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/evaluate.py
View File

@ -122,7 +122,7 @@ def evaluate(
checkpoint_dir=checkpoint_dir, checkpoint_dir=checkpoint_dir,
template=template template=template
)) ))
chat_model.tokenizer.padding_side = "left" # avoid overflow issue in batched inference for llama2 chat_model.tokenizer.padding_side = "right" # avoid overflow issue in batched inference for llama2
eval_template = eval_templates[lang] eval_template = eval_templates[lang]
category_corrects: Dict[str, np.ndarray] = { category_corrects: Dict[str, np.ndarray] = {