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

Why multiply flow ground truth by 0.05? #11

Closed
dsqx71 opened this issue May 12, 2017 · 2 comments
Closed

Why multiply flow ground truth by 0.05? #11

dsqx71 opened this issue May 12, 2017 · 2 comments

Comments

@dsqx71
Copy link

dsqx71 commented May 12, 2017

I was wondering why you multiply ground truth by 0.05. It seems to influence actual learning rate.

@eddy-ilg
Copy link
Collaborator

It is not equivalent to modifying the learning rate. The reason we did it is to have image and flow values normalized to a range of approximately -0.5 to 0.5. It seems that it is not absolutely necessary though.

@dsqx71
Copy link
Author

dsqx71 commented Jun 1, 2017

Thanks for your reponse. I got your point.

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