Why the ids list for each task is scaled by a factor, e.g., self.rs_ids = self.rs_ids * 120

va1shn9v / PromptIR

PromptIR: Prompting for All-in-One Blind Image Restoration [NeurIPS 2023]

https://arxiv.org/abs/2306.13090

Other

297 stars 22 forks source link

Closed xl-tang3 closed 3 months ago

xl-tang3 commented 4 months ago

what is the purpose of this scaling and how it will affect the training results?

shallweiwei commented 3 months ago

I believe this is to solve the problem of imbalanced data