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.

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

Implement IOU loss instead of cross entropy loss #3

Open
abfarahani opened this issue Feb 23, 2019 · 2 comments
Open

Implement IOU loss instead of cross entropy loss #3

abfarahani opened this issue Feb 23, 2019 · 2 comments
Assignees
@jiahaoxu @abfarahani
Labels
todo Jobs that are going to be done

Comments

@abfarahani
Copy link
Contributor

Try to implement the IOU loss, and compare with the cross entropy loss
@abfarahani abfarahani added the todo Jobs that are going to be done label Feb 23, 2019
@jiahaoxu
Copy link
Contributor

The dice loss makes the model predict all background ('0').
Also, we are going to try focal loss as well.

@abfarahani
Copy link
Contributor Author

We tried both dice loss and focal loss, but none of them worked. We may work on them as future works.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@jiahaoxu jiahaoxu

@abfarahani abfarahani

Labels
todo Jobs that are going to be done
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@jiahaoxu @abfarahani