you need cuda 12.4, python 3.10 and cudnn 8.9.7 and c++ and visual studio
you need cuda 12.4, python 3.10 and cudnn 8.9.7 and c++ and visual studio
FYouTubeSECourses
If you are interested in using AI, generative AI applications, open source applications in your computer, than this is the most fundamental and important tutorial that you need. In this tutorial I show and explain how to properly install appropriate Python versions accurately, how to switch between different Python versions, how to install diffe...
IHi everyone, should I use regularisation images when training dreambooth for sdxl?
DHi, @Dr. Furkan Gözükara Did you try training OpenFLUX / Schnell with kohya's latest commits on sd3-flux.1 branch ? I always get a NotImplemented error. tried on torch 2.4.1 and 2.5.0
Dtried a diffusers format and bfl too
Danyway just wanted to know if its familiar to you
SDr Furkan i hope everythong is ok, i just saw thta onetrainer support Flux , do you plan to do a tutorial in onetrainer sometime in the future?
PDr. how to set Kohya SS, in a DreamBooth FLUX Train, to save the train every "x" steps if I need to stop the train and restart it before exactly where I stopped?
Xgot live portrait to work!
Xsuper good video going to try experimenting with other ai apps soon after watching the video i feel confident
VWhen fine tuning flux, any suggestions on lora_rank? The default on the replicate trainer is 16, and it says higher numbers capture more features but take longer to train. Any guidance on this?
XXHaven’t tried it
Fi tried official Schnell it was working
F23.8 gb
Fyes but they still dont have fp8
Fi am waiting upgrade of that
Fi already have some configs working
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
Fyou have to start a new training
Fit is same as lora
Fsave every n steps
Fwe use 128
Fi compared lora rank impact
Fmultiplem times
KI'm ways from understanding the logic of Kohya training. To many parameters. I can't for the life of me understand why there is no better way to do it than the folder preparation 
And I have watched your video several times. It's a looooong video 
And I have watched your video several times. It's a looooong video KIn the video you say "You need to select LORA. Because we are currently training LORA"... Which we aren't since now we're talking finetuning, so it's a little bit confusing
Fyes now select dreambooth :d
Fso there are 2 trainings we do
F1 lora
F1 dreambooth
Fi will also hopefully make a dreambooth tutorial
Fperfecting it
Ftesting 15 imgs batch size 7 atm
F
KNo it's not. Use symbolic links! This makes it easier -> https://schinagl.priv.at/nt/hardlinkshellext/linkshellextension.html
KBS 7 is way better
Fi will compare it with bs 1
Fboth 15 images
Fboth 256 images
Fby the way i researched a unique LR for bs7 :d
KLooking at the right column it's way better than the left column
KI need to get into the "Kaggle" thing I guess. I can't keep locking up my computer
Fthese are all same bs7
Fdifferent checkpoints
Ffinding best epopch
KRan 2 trainings without a problem with AI-Toolkit over night though.. Unfortunately bad ones (As expected)
KSorry read it wrong! Read BS 1 for the left column
F15 images bs7 really good
