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

Training Cityscapes dataset #76

Open
CurryKd7 opened this issue Sep 25, 2024 · 1 comment
Open

Training Cityscapes dataset #76

CurryKd7 opened this issue Sep 25, 2024 · 1 comment

Comments

@CurryKd7
Copy link

Why is this link https://www.rocq.inria.fr/cluster-willow/rstrudel/segmenter/checkpoints/cityscapes/seg_large_mask/log.txt The mlou of the training citycpales dataset in the first round can reach 43%, but I only trained for 9% in the first round, and I used vit_1arge_patch16_384 as the backbone. May I ask why this is?

@LUPIN11
Copy link

LUPIN11 commented Sep 25, 2024

Maybe the defaulted settings (layers, lr, batchsize) are changed on this variant. I also reproduced this variant on a single gpu, the default settings should be modified.

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