Skip to content

训练CRNN #30

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
Open
Jimmy-Cheong opened this issue Apr 15, 2020 · 5 comments
Open

训练CRNN #30

Jimmy-Cheong opened this issue Apr 15, 2020 · 5 comments

Comments

@Jimmy-Cheong
Copy link

作者您好，想请教一下，我用自己合成的数据集训练CRNN，loss降下来了，但是拿图片测试不出结果，大概是哪里出了问题呢
@courao
Copy link
Owner

courao commented Apr 15, 2020

你好，
请问你是用合成的数据训练然后其他的数据测试吗？
如果训练集测试集差异过大的话可能会出现这样的情况。

@Jimmy-Cheong
Copy link
Author

@courao 对的，那请问这个有什么解决办法吗

@courao
Copy link
Owner

courao commented Apr 16, 2020

最好是用一些贴近测试数据的图片来训练

@Jimmy-Cheong
Copy link
Author

@courao 我用了相近的图片测试效果是可以的，那就是数据集太单一了，请问您用的是什么数据集呢

@courao
Copy link
Owner

courao commented Apr 16, 2020

我用的是合成的数据加上一部分学校里项目用到的数据

@Jimmy-Cheong
Copy link
Author

好的，谢谢您

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
@Jimmy-Cheong @courao