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

训练时的验证集精度与评估的精度不一样。(评估的也是相同验证集) #141

Open
DIE35 opened this issue Jan 18, 2025 · 2 comments

Comments

@DIE35
Copy link

DIE35 commented Jan 18, 2025

up主你好,我在使用mobilevit的时候遇到以下的问题:
首先我训练时候的验证集最佳精度为89%,然后我使用evaluation.py去评估验证最高精度的权重,但是结果比训练的验证集最佳精度低2个点,但是评估和训练我都用的一个验证集,这个是什么问题呢 ?
而且这种低多少或者高多少对于不同的训练模型,都是不固定的,如果这样的话,我对模型进行改进,没有办法准确的评估数值呢。

@Fafa-DL
Copy link
Owner

Fafa-DL commented Jan 19, 2025

应该是验证和训练时设置的随机种子导致的

@DIE35
Copy link
Author

DIE35 commented Jan 20, 2025

应该是验证和训练时设置的随机种子导致的

up主,这个有办法解决吗?能够使得评估和测试的精度保持一致,随机种子我固定了后,发现评估精度更低了。

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