Skip to content
This repository has been archived by the owner on Jan 1, 2025. It is now read-only.

training from scratch, loss decreased very slow #137

Open
chrisx599 opened this issue Aug 31, 2024 · 1 comment
Open

training from scratch, loss decreased very slow #137

chrisx599 opened this issue Aug 31, 2024 · 1 comment

Comments

@chrisx599
Copy link

have anyone who training from scratch, not use pre-trained weight of ViT-B?
Can author or anyone released the training log of train from scratch?
the author said it takes more epochs to training from scratch, but did'nt released a concrete number, can author says this number?
when i trained 15 epochs, the accuracy is only 16%, and from the beginning, it's just 4%, i'm so confused is that normal?
image
image

@chrisx599
Copy link
Author

first figure is loss, seconde figure is accuracy

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant