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

Hello, I find your work very interesting and have a question for you! #13

Open
Minsky520 opened this issue Mar 6, 2024 · 1 comment

Comments

@Minsky520
Copy link

how did you determine that the scaling factor is equal to 60? Does this have any significance?

@songrise
Copy link
Owner

songrise commented Mar 20, 2024

Sorry for the delay. This is a good question, the scale of 60 is directly adopted from the previous paper, CounTR (example in this line).

Basically, it will scale up the gt by 60, and when calculating the metric the scale will be canceled out. Therefore, we believe it will not affect the performance as long as the same scale is applied for training and inference (and there is no numerical instability). However, we did not experiment with other scales and you may try how different scale may affect the model.

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