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.

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

develop版本,在P800上,使用PPSeg的OCRNet,MixLoss选择DiceLoss和LovaszSoftmaxLoss会报错 #70805

Open
trarynight opened this issue Jan 13, 2025 · 5 comments

Comments

@trarynight
Copy link

trarynight commented Jan 13, 2025

bug描述 Describe the Bug

config配置如下:
Loss:
types:
- type: MixedLoss
losses:
- type: DiceLoss
- type: CrossEntropyLoss
- type: LovaszSoftmaxLoss
coef: [0.1, 0.6, 0.3]
- type: MixedLoss
losses:
- type: DiceLoss
- type: CrossEntropyLoss
- type: LovaszSoftmaxLoss
coef: [0.1, 0.6, 0.3]
coef: [1, 0.7]
报错截图如下:
/
error_250113

其他补充信息 Additional Supplementary Information

No response

@deepllz
Copy link
Contributor

deepllz commented Jan 13, 2025

你好,请问相同版本的A卡,或者H卡上,会报错吗?

@trarynight
Copy link
Author

我们在A100和910B都做过训练适配,不会有训练报错情况。

@deepllz
Copy link
Contributor

deepllz commented Jan 13, 2025

请问是场内的员工嘛?可以私信,我拉一下p800的同事帮忙看一下,可能是版本问题

@deepllz
Copy link
Contributor

deepllz commented Jan 13, 2025

你好,可以详细写一下如何复现这个问题,主要是环境部署,paddle的commit_id,以及启动命令。后续找相关同事排查一下。

@trarynight
Copy link
Author

是,之前也找了昆仑的同学帮忙,问题一直未解决。我如流私信你

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants