Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

找不到对应模块 #13

Open
gogelan opened this issue Nov 10, 2024 · 5 comments
Open

找不到对应模块 #13

gogelan opened this issue Nov 10, 2024 · 5 comments

Comments

@gogelan
Copy link

gogelan commented Nov 10, 2024

您好，非常感谢您杰出的工作！
我在执行bash experiment/mimic3/online_distill.bash命令时报错：

File "/home/Users/LEADER-pytorch/models/bert_models.py", line 17, in <module>
    from models.graph_models import FuseEmbeddings
ModuleNotFoundError: No module named 'models.graph_models'

原因好像是models文件夹下没有graph_models模块，请问可以上传对应的代码吗？
@ricartojason
Copy link

请问一下你运行成功了吗？那个llama-7b能不能用llama2代替呀？

@gogelan
Copy link
Author

gogelan commented Nov 11, 2024

您好！

  1. 我在llama-7b上跑通了bash experiments/llm_cls.bash命令，也得到相关结果数据，但是在知识蒸馏训练小模型的过程中报错，相关错误如一楼所述。

  2. 我在运行该项目的过程尝试过llama2-7b-hf，训练到46%报错：

Exception: Current loss scale already at minimum - cannot decrease scale anymore. Exiting run.

建议先用llama-7b跑通试试，如果您尝试llama2的过程中也遇到相同情况烦请告知我解决办法，谢谢！

@ricartojason
Copy link

好的，感谢您的答复！另外请问一下你的依赖文件是完全参考requirements.txt的吗？我按照requiements.txt安装了全部依赖，且cuda、torch都是正常运行，但是我在运行大模型时
bash experiment/llm_cls.bash
遇到了如下错误：
RuntimeError: ProcessGroupNCCL is only supported with GPUs, no GPUs found!

@gogelan
Copy link
Author

gogelan commented Nov 12, 2024

这种错误应该是cuda之类的没有设置好，建议根据报错信息上网找一下解决方案。

@ricartojason
Copy link

非常感谢您的回复！不知道能否请求一下您的邮箱，想查看一下您的相关依赖库版本可以吗？

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@ricartojason @gogelan