neko
neko
或许你可以尝试使用CAIN而不用cain-ncnn-vulkan
Doesn't the tilesize have to be divisible by 32?
I trained a new CAIN model, but I don’t know how to convert the pytorch model to the ncnn model
I used CAIN to make up a video, here is the link: https://drive.google.com/file/d/1w-4oU9e-LSU2JSCSbulX0Dueqdiv2Czc/view?usp=sharing
It's a bit difficult for me ε(┬┬﹏┬┬)3
https://drive.google.com/file/d/1gazwEvXVVEzAwSORoMzPhihj1ylfQzVN/view?usp=sharing
vimeo90k 300epoch https://drive.google.com/file/d/1yXISQLCr15LU0pDzsZMrZQzSLHJgyz1K/view?usp=sharing
Have you tried -t 2048
I think this should be a problem with the CAIN algorithm
It is recommended to use rife-ncnn-vulkan instead of cain-ncnn-vulkan