Double Feature: Going for Gold
Added 2023-01-03 22:40:13 +0000 UTC(obligatory non-milk version for later DeviantArt upload included)
Turns out training an AI to label the data to train another AI is turning into a real hassle. It's going well, but still a totally different world. Straying a single inch off the beaten path of Automatic1111's walled garden you quickly realize how firmly he was holding your hand: the world of AI is so much more convoluted, complicated and deep than just diffusion, let alone the made-to-order diffusion UI we've been given on a silver platter
Hopefully, though: the effort will be worth it. Like all things in computer science: the diffusion finetune, dreambooths and the like follow the rule of 'garbage in, garbage out', which also means enriching your data to be more accurately captioned should have a meaningful effect on the resulting models' performance and flexibility
Comments
quite a bit, none of these are one-click-wonders (most generated stuff that's of any quality usually isnt; despite what many may believe)
Bury She
2023-01-11 06:32:03 +0000 UTCGod these are amazing! How long did it take you to make these?
Goomba
2023-01-04 06:20:42 +0000 UTCExtremely plump 💙💙💙
Creepette
2023-01-03 23:02:04 +0000 UTCWow! AI Stable Diffusion is the best. Keep on the future.
JomLagr
2023-01-03 22:45:32 +0000 UTC
