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

About metric learning output embedding #4

Closed
ElegantLee opened this issue Nov 11, 2021 · 3 comments
Closed

About metric learning output embedding #4

ElegantLee opened this issue Nov 11, 2021 · 3 comments

Comments

@ElegantLee
Copy link

作者您好,论文中您在Unconditional Image Generation和Single Image Super-Resolution这两个任务上为netML设置了不同的output embedding(分别为10和32),如何根据自己的任务来选择合适的output embedding呢?比如Image-to-Image translation这样的任务。

@dzld00
Copy link
Owner

dzld00 commented Nov 11, 2021

这个问题我们也没有确切的答案,在一定范围内应该都是合理的。我们的一个经验是对于分布复杂的数据用高一些的维度可能更有利,对于分布简单的数据则不需要

@ElegantLee
Copy link
Author

这个问题我们也没有确切的答案,在一定范围内应该都是合理的。我们的一个经验是对于分布复杂的数据用高一些的维度可能更有利,对于分布简单的数据则不需要

好的,感谢。另外,使用MVM进行训练,流行匹配的loss是下降的,度量学习部分的loss呈上升趋势,这是否意味着训练失败了?
image

@dzld00
Copy link
Owner

dzld00 commented Nov 12, 2021

训练稳定的话mm_loss的确在下降,ml_loss在一定区间内振荡,如果不是的话训练可能就失败了,不过这是稳定训练的必要条件不是充分条件

@dzld00 dzld00 closed this as completed Nov 16, 2021
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