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 mlp_ratio in the SR x2 #7

Closed
liboyun opened this issue Mar 25, 2024 · 2 comments
Closed

About the mlp_ratio in the SR x2 #7

liboyun opened this issue Mar 25, 2024 · 2 comments

Comments

@liboyun
Copy link

liboyun commented Mar 25, 2024

Thank you very much for your work! I am curious about the mlp_ratio in the SR x2 settings.

In the options/train/train_MambaIR_SR_x2.yml, the mlp_ratio is 2.0, while in the log provided in here is 4.0.

I am wondering which one should we tend to refer? Expect for you reply.

@csguoh
Copy link
Owner

csguoh commented Mar 25, 2024

Hi, the mlp_ratio in the log file denotes the MLP ratio in the Transformer inherited form previous code base, which means it is a no-use hyper-parameter. In the released yml file, the mlp_ratio represent the hidden-space expansion ratio in the SSM, you can use the default value mlp_ratio=2 to reproduce the experimental results ;D

@liboyun
Copy link
Author

liboyun commented Mar 25, 2024

Thanks for your replies. 👯

@liboyun liboyun closed this as completed Mar 25, 2024
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