search

98mxr / GMFSS_union

High Performance GMFSS with RIFE and GAN
MIT License
24 stars 4 forks source link

FP16 doesn't work #2

Open korakoe opened 1 year ago

korakoe commented 1 year ago

using the --fp16 command results in this error: RuntimeError: Input type (float) and bias type (struct c10::Half) should be the same

98mxr commented 1 year ago

I'm very sorry about this, I just inherited inference_video.py, and didn't maintain all its functions on GMFSS, it is known that --montage and --fp16 should not work properly, --UHD is also de facto obsolete state. And I have doubts about the effect of using FP16 to infer GMF. If I just want to try GMFSS_union under 8G video memory, I can add a mixed precision modification to do this.