Lora微调完Llama-3-8B-Instruct拿来预测，does not appear to have a file named config.json - Githubissues

hiyouga / LLaMA-Factory

Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

https://arxiv.org/abs/2403.13372

Apache License 2.0

34.19k stars 4.21k forks source link

Lora微调完Llama-3-8B-Instruct拿来预测，does not appear to have a file named config.json #5989

Closed wutianhao666 closed 2 days ago

wutianhao666 commented 2 days ago

Reminder

[X] I have read the README and searched the existing issues.

System Info

llamafactory version: 0.9.1.dev0
Platform: Linux-5.15.0-122-generic-x86_64-with-glibc2.31
Python version: 3.12.4
PyTorch version: 2.4.1+cu121 (GPU)
Transformers version: 4.44.1
Datasets version: 2.19.1
Accelerate version: 0.34.2
PEFT version: 0.12.0
TRL version: 0.9.6
GPU type: NVIDIA GeForce RTX 4090
vLLM version: 0.6.1.post2

Reproduction

Expected behavior

No response

Others

Lora微调完保存模型的路径为LLaMA-Factory/saves/Llama-3-8B-Instruct/lora/train_2024-11-10-19-46-32，上图是该目录下的全部文件，确实没有config.json。我在利用这个微调结束后的模型进行生成，报错如上图

hiyouga commented 2 days ago

https://github.com/hiyouga/LLaMA-Factory/blob/main/examples/train_lora/llama3_lora_predict.yaml