how could I replicate this?
#3
by
Meeeeq
- opened
hello, may I have some guidance to how this works?
does that dataset you use "6.5" mean 6.5 billion images or million? The images all seemed smaller than 256x256?
Would this model do better than a dreambooth trained model trained on cropped 256x256 images? I believe it does.
I would like to train a model like this too.
Thanks
@justinpinkney
!