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

数据集小导致boost无法获得伪标签 #17

Open
Wt197012 opened this issue Jun 7, 2023 · 4 comments
Open

数据集小导致boost无法获得伪标签 #17

Wt197012 opened this issue Jun 7, 2023 · 4 comments

Comments

@Wt197012
Copy link

Wt197012 commented Jun 7, 2023

尊敬的李博士,我正在尝试用自己的数据集进行聚类,但我发现我的数据集比如说两类各100张图时,训练过程正常,但在boost过程中无法获得伪标签导致报错,请问这种情况可以通过修改网络解决吗?谢谢!

@Yunfan-Li
Copy link
Owner

您好,可以根据筛选情况尝试调整伪标签的比例和阈值这两个超参数

@Wt197012
Copy link
Author

Wt197012 commented Jun 7, 2023

好的,谢谢您!

@Wt197012
Copy link
Author

Wt197012 commented Jun 7, 2023

您好,我现在的boost过程中准确率在逐步下降,请问时通过增加alpha和减小gamma来调试吗?谢谢!

@Yunfan-Li
Copy link
Owner

可以查看一下伪标签的准确率是否始终高于整体的准确率,然后伪标签类别是否平衡,可以通过增加alpha和减小gamma来尽量提升伪标签的质量。

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

No branches or pull requests

2 participants