Half precision (fp16) Training and Inference of the HAT models

[AmirMohamadBabaee]

Hello,
Thank you for sharing your fantastic work.
I'm curious if you have any intentions to share the training and inference configurations using fp16 to accommodate limited GPU memories.

[chxy95]

@AmirMohamadBabaee Haven't tried it yet~

[AmirMohamadBabaee]

If such plans exist, I kindly request further details regarding these configurations. However, should this not be part of your current roadmap, I would greatly value your insights into how I might implement it independently.

[chxy95]

@AmirMohamadBabaee To be honest, I'm inexperienced in doing this...