There is something wrong, please query again

[karry5921]

运行app.py脚本时，出现了这个错误：There is something wrong, please query again。 报错信息如下：

微调后生成的文件在截图里所示的目录下，那么，运行app.py时，model_name_or_path的值如何指向微调后的模型，这个值应该是什么呢？ 那如果是lora微调，情况又是怎么样呢？

[research4pan]

Thanks for your interest in LMFlow! If you would like to use a finetuned model, you can simply change line 27 to model_name_or_path = path-to-your-finetuned-model, which specifies the path to the finetuned model. To load lora model, you may specifies

lora_model_path = path-to-your-finetuned-lora
model_args = ModelArguments(model_name_or_path=model_name_or_path, lora_model_path=lora_model_path)

Hope that can solve your question. Thanks 😄

[SHUMKASHUN]

Hi, For the first question, it seems you are running an old version of LMFlow, where app.py may not support multi GPUs serving, and you are using multi GPUs. You can either try specify CUDA_VISIBLE_DEVICES=0 to only one device, or update to our latest code to run multi GPUs setting.

[karry5921]

Got it ! Thank you !

[karry5921]

Hi ! I solved above issuses, but there is another question. The finetuned model is still unable to give the correct answer. I can't figured out what caused this problem . Could you please give me some suggestion?

[shizhediao]

The model you are using is not good. You might want to try our released model first. https://github.com/OptimalScale/LMFlow#53-reproduce-the-result

[shizhediao]

This issue has been marked as stale because it has not had recent activity. If you think this still needs to be addressed please feel free to reopen this issue. Thanks