search

SHI-Labs / Smooth-Diffusion

Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Models arXiv 2023 / CVPR 2024
https://shi-labs.github.io/Smooth-Diffusion/
MIT License
319 stars 8 forks source link

What is the expected range of reg_loss? #16

Open Nayoung-Oh opened 3 months ago

Nayoung-Oh commented 3 months ago

Hello, thank you for sharing good work! I tried to apply the reg_loss calculated by the step_regularize() function in train_smooth_diffusion.py in other diffusion models. However, I found out that the reg_loss shakes a lot (1 ~ 100). What is the expected range of reg_loss? Or, is there any additional skill necessary to be applied?

Thanks:)