@Furkan Gözükara SECourses do you think dreambooth is better then hypernetwork training in web ui?
@Furkan Gözükara SECourses do you think dreambooth is better then hypernetwork training in web ui?
Sdreambooth will give me a checkpoint right?
FI think dreambooth extension has matured greatly
FYes 4gb safe tensors file
Shmm yeah but then i wont be able to create my characters in different models
FWith lora you still won't be very much
FEach training only works best on trained model
Shmmm but I've tried different loras on multiple models and they all give a great result and now with the new update to lora which is loha
Sthe results r amazing
Sand even half the size
Fthey must have common root model probably
Syeah ofc the sd1.5
Fso you tried default settings? i plan a tutorial but still couldnt find time yet
Soh yeah i can do normal lora training and the results r good but the loha one is new and I cant find many tutorials since it just came out recently but the ppl who have made some r giving great results even with default settings
SGitHub
Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion. - GitHub - KohakuBlueleaf/LyCORIS: Lora beYond Conventional methods, Other Rank adaptation Implementati...
Fi will ask developers
Fok extended lora in dreambooth extension of web ui already supports these
II tried to replicate your last settings today, and for the first time I got complete noisy images.
II'm using your exact same specifications, same GPU, and almost identical photos for learning purpose.
IXfomers removed for me as well. Although the model shows images on Runpod, it shows only noise on Automatic 1111
IIn general, I don't understand how you trained your models, because with the same exact settings as in your videos (and I mean EXACT), I don't even get a close resemblance to my photos!!!
IIt's frustrating, and we are spending too much money on Runpod, and got zero models to use.
Fare you using yaml file it generated?
IYes copy/paste in models' folder
Fif it works on runpod and not on your pc
Fdefinitely something not right on your pc
Fnoise usually happens due to some incorrect installation or configuration
IThis only happens when used "Lion"
Fi am using lion and just works fine for me. actually i tested trained models on my pc too
Fworks perfectly fine
Fit should also work. because we are using same code both on pc and runpod
Fhave you tried with fresh installation?
IYes, I'm using the latest installation after the last update.
IMy still shows xformers though!
I*Mine
Fdid you try with torch version 1?
Fperhaps torch 2 issue?
Fwhich xformers do you see on your web ui
IFollowing your last video, torch 2 and everything, worked perfectly well for me.
IOne problem, it wouldn't start offline.
Ftorch 2 is still in development
Fso if something not working right
Fi think make a new folder
Finstall there torch 1
Fand test there
IAnd today this happened with only the models trained using "Lion"
Fhmm
Fmaybe torch 2 not compatiable i dont know lion shouldnt make such difference because it is just an algorithm during training
IBut also the Web UI needing an Internet connection, which wasn't necessary before the last update
Finteresting. that means it is doing some update