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 the loss function l1 loss and ae loss #42

Closed
kikyou123 opened this issue May 23, 2018 · 3 comments
Closed

about the loss function l1 loss and ae loss #42

kikyou123 opened this issue May 23, 2018 · 3 comments

Comments

@kikyou123
Copy link

In your paper, I can only find the l1 loss, I am confused about the ae loss?

@JiahuiYu
Copy link
Owner

In code, we use l1 loss to regress only the masked regions, and use ae loss to regress the unmasked regions. This simply means l1 + ae = ae loss of entire image. The paper's description is correct.

@wuyxiquanquan
Copy link

Where does the AE loss come from?

@buchholzmd
Copy link

@wuyxiquanquan this is the auto-encoder loss (as in BEGANs). The l1 loss regresses the holes (masked regions) to the ground truth regions. The ae loss is the standard auto-encoder loss to reconstruct the rest of the image. Note that the Generator loss is l1 + ae, as described above.

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

4 participants