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

Loss does not decrease for coco #36

Open
strongwolf opened this issue Aug 30, 2018 · 3 comments
Open

Loss does not decrease for coco #36

strongwolf opened this issue Aug 30, 2018 · 3 comments

Comments

@strongwolf
Copy link

Have you ever trained on coco for these models? I try to train coco with original SSD. However, the loss keeps increasing especially for the localization loss from 3 to 30! The accuracy also decreases. I use the default config and everything behaves well on Pascal voc.

@lzx1413
Copy link
Owner

lzx1413 commented Aug 30, 2018

I will try it later. Please keep pending.

@Chida15
Copy link

Chida15 commented Aug 1, 2019

Have you ever trained on coco for these models? I try to train coco with original SSD. However, the loss keeps increasing especially for the localization loss from 3 to 30! The accuracy also decreases. I use the default config and everything behaves well on Pascal voc.

Hello, I also meet this problem, have you solved it?

@strongwolf
Copy link
Author

Have you ever trained on coco for these models? I try to train coco with original SSD. However, the loss keeps increasing especially for the localization loss from 3 to 30! The accuracy also decreases. I use the default config and everything behaves well on Pascal voc.

Hello, I also meet this problem, have you solved it?

The reason is that the augmentation operations have some errors that are not appropriate for some background images in coco.

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

3 participants