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

OSpatialNet貌似出现过拟合的问题？ #24

Open
A-little-star opened this issue Apr 3, 2024 · 3 comments
Open

OSpatialNet貌似出现过拟合的问题？ #24

A-little-star opened this issue Apr 3, 2024 · 3 comments

Comments

@A-little-star
Copy link

你好，我用开源的SpatialNet和OSpatialNet分别训练了语音分离的模型，SpatialNet的表现确实非常惊艳，但是OSpatialNet出现训练时loss下降的比较正常，但是测试集中得到的结果非常差。猜测可能是出现过拟合的问题？
@quancs
Copy link
Member

quancs commented Apr 3, 2024 •
edited

您好，感谢关注我们的工作。对于oSpatialNet，我们主要遇到的主要问题是长度外推问题。泛化问题我们没有遇到，您这个是不是泛化问题需要进一步研究、分析

@A-little-star
Copy link
Author

好的，我再去研究研究

@A-little-star
Copy link
Author

好吧，其实并不是过拟合的问题，而是训练和推理时模型的超参数不匹配导致的。训练时采用的MultiheadAttention的num_heads为4，而推理时采用的num_heads却是2，从而导致的推理结果差。。。

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
@quancs @A-little-star