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

关于Chinese-CLIP复现中的几点疑惑,期待答复! #91

Open
kongyan66 opened this issue Mar 19, 2024 · 0 comments
Open

关于Chinese-CLIP复现中的几点疑惑,期待答复! #91

kongyan66 opened this issue Mar 19, 2024 · 0 comments

Comments

@kongyan66
Copy link

kongyan66 commented Mar 19, 2024

作者您好,对们的工作十分感兴趣,我有几个关于Chinese-CLIP的疑惑,还请解答:

  1. 关于CTR阶段的模型权重是否可以提供呢?我想对齐下论文的结果。
  2. 关于单字识别,作者epoch训练的设置为多少,我看默认10000, loss大概的范围才是收敛呢
    我使用HWDB1.0 训练99个epoch,在HWDB1.0 测试WACC大约88%,测试集说用的是ICDAR2013并没有找到单字的,网上只有单行的,所以表二结果无法对齐,请问是否可以提供下载链接。
  3. 关于行文本识别,训练数据用的哪些,我使用HWDB2.0-line,发现训练数据中有符号(,。),而CCR-CLIP模板中不支持的符号,强加的话这个也得重新训练,请问这个如何解决呢?
  4. 如果想支持符号或者英文字符,CLIP的test encoder应如何构建呢?
    期待您的答复!
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

1 participant