only fixing the human class will be a success
only fixing the human class will be a success
Cand IF they want special loras, he can deal with that for them himself
Ctrust me on this, okay? he tried. that's why it's so overfit for women
Cor, well, maybe it's just heavily biased for women. same effect however. here's the phrase "Pastel pencil". two women, a dog, and a rose
Cno landscapes, or anything else. 3 out of 4 are women and dogs
Cwith an anime look to one of the women
Cand i jsut generated that
PAhh nice! Thankls for the tip 
Cand the thing that using only "Pastel pencil" SHOULD have returned is - 4 images of actual pencils
Cyou don't see pencils at all
Dyou can train with a very diverse people all genders all races all ages 3000 pictures at least for many epochs and may be end up more flexible and training on top produces lees bleeding, that will be a success for me, idk
Cthat would require flux to be trainable. it isn't.
Cnot even robin can train it. he'd have to train the base model, and that's expensive. VERY expensive. hundreds of dollars at the minimum, but more like thousands of dollars if not more
Ci was seriou when i said that training the base model was at least a million dollars
Cand for him to train new data into it, that's not making a lora, that's running the training on the base model on super clusters of GPUS for quite a while.
Dyes but training not the entire model, just the human class
Dthat will be cheaper
Cthat requires making a lora. that's what loras are for. and that can not be done
Che would have to train the base model and update the weights in it
Dyou can do a dreambooth
Cand IF he has a client, like a hollywood studio, that's willing to foot the bill and pay him a profit, he'll do it for them
Dand a new checkpoint
Cbut that'll be pro only, and probably an exclusive model
Cit won't be availble to anyone else.
Cno, you can't. dream booth doesn't allow you to train the entire base model. you are training a checkpoint but that isn't something you can do with flux. you have to train the entire base model.
Cthe neural network he decided to use is a bloody LLM! for crying out loud
Dok, I get your point
Dthanks
Chere's a silly prompt: a tired, weary man sitting in a lawn chair on the beach. A large crab looms behind him, its claws spread open in a threatening manner, and in the distance we see a whale spouting water into the air
DThanks man, I'm going to sleep I have to work tomorrow
Cnight
CFantastic thank you very much!
Dsaw an interesting comment. are different blocks responsible for face likeness and overall image look? or its just broscience?
CEach block does something specific
Fyes more memory better. i plan to upgrade 128gb
Ftrue but i am planning models to get more ready
Fstill too early i think
Cyou need vram, not system ram
Fok i see his method seems very weak
Fno i doubt that it will ever work 
Fi need both :d
Cnot really. you shouldn't be using that CPU for anything graphics related. and if you have an AMD, you should go through CS1o's AMD guides and make sure it's set up right.
Ceverything should be on your GPU and in vram
Fit is pretty much nonsense
Factually i was expecting training single block would reduce bleeding but it didnt
Fi guess due to transformer arcihecture even single layer changes alls output
Fit depends on how many apps you open at the same time
Cthere's a 12 step program for that
X@Dr. Furkan Gözükara have you tried flux 1.1?
Fit is only api