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

Vector Quantization loss rises in training #219

Closed
OrangeSodahub opened this issue Sep 1, 2023 · 3 comments
Closed

Vector Quantization loss rises in training #219

OrangeSodahub opened this issue Sep 1, 2023 · 3 comments

Comments

@OrangeSodahub
Copy link

Hi, guys, I may want to ask for help with the training of a VQModel from scratch, while in my case, the vq loss rises in the middle of training process, while the reconstruction loss kept decrease, the vq loss kept rises to a large level.
I'm really confused about it, is there anyone know how to solve it or ever encountered this issue?
@OrangeSodahub
Copy link
Author

Solved.

@jasong-ovo
Copy link

I meet the same phenomenon. How did you solve it ?

@OrangeSodahub
Copy link
Author

You could check if there is something wrong with the gradient of each parameter of your model first

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@OrangeSodahub @jasong-ovo