@Dr. Furkan Gözükara have you tried reg images with dreambooth?
@Dr. Furkan Gözükara have you tried reg images with dreambooth?
Fyes i am testing training results now
Fif you mean fine tuning
Fi think i tried didnt work
Fi am testing onetrainer fix
Fthing
Myeah for dreambooth
Montrainer one i tried like 7 different trainings, couldnt get it to work
Fye me neither
Fnot working
F
MFurnald Gozump
Fcould cause i added text encoders?
F
Fi didnt train though
Mhe said he wasnt training them
Fi also dont know what is meaning of include
Mon onetrainer you need to include them either way
Fit doesnt make sense
Mwhen you have the include option its like in kohya when you load in clip l and t5
Fphoto of a man
Myep had the same issue with mine
Mi wish @dxqbYD could provide a detailed workflow and training so we can see it properly
Dincluding TE Is fine just no training
Dlet me know how I can help, but I currently don't know what you could do differently. if you look at my samples on lower left and lower right, the images are nearly identical. dreambooth-training
Dyou could provide config and images and I'll have a look? but I'm on a business trip from Mo-Wed
Myou could provide info like, learning rates, optimizers, noise offset etc
DI could, I just have no reason to assume that is what makes the difference - because the difference is not just a little, but night and day
D
Dnevermind the "train TE1 embedding" switch, that's just something I'm currently trynig. my samples were without
Dalso, LR is not comparable to yours because I am using alpha 1
DI actually don't think this will be useful in the end, because it limits training of the concept you want to train (or 'instance' as kohya would call it) too much. but that's the opposite of the problem of what you're seeing
Dthat enough?
Moh wow your learning rate is super high
Dno it's not. it's because I'm using alpha 1.
Dif you are using 16 you can practically divide the LR by 16
Dwhat LR and alpha are you using?
M128x128 and 3e-06
Dthen my LR is approximately equivalent to 8e-06 in your world
Dso 3x but Flux is quite tolerant on a wide range of LR. I don't expect that to be the difference
DI think it's gonna be more productive if one of you just sends me their dataset and config, and I'll see if I can make it work.
Dit's not like Furkan's face is a secret, it's all over the internet 
Dso send me the 16 pics + config and I'll try
B@Dr. Furkan Gözükara which of your Flux Lora training configs is a good balanece of speed with true photo realism?
Bim trying to get the trainer on fal.ai to get as good results in terms of detail and realism as your kohya configs. the fast trainer doesn't do it very well so im using the regular trainer.
Fboth fast and good on 24 gb gpus
BThank you
Dusing embeddings or TE training might solve that issue though. Not bad on the left with embedding, on the right just "woman" with no bleeding.
FTE training doesnt solve
Ehey @Dr. Furkan Gözükara quick question. Can Flux 1.1 be Lora trained or even downloaded? As far as I see it can only be used in platforms at the moment.