How long should using 20 images take to train using the 24gb json file roughly?
How long should using 20 images take to train using the 24gb json file roughly?
Fdepends on gpu
A3090 is what I have
Fbut lets say you did 150 repeat
Fye around 1.7 second / it lets say
Fso
F6000 * 1.7 second
Ain your latest post there what did you do change the 24gb json file configs?
Flearning rates
Fread the patreon post very detailed
Fi should test and see if SD 1.5 broken too or not :/
DI'm training on an rtx 3090 with Gradient checkpointing off and CrossAttention sdpa at 1.52s/it
Fnice speed
Fspda speed up
Fbut i dont know how much quality loss
D1.51s/it now, still slower than previous versions with the same settings but it is faster than using Gradient checkpointing on this capture is with an older version
DI didn't know that this option can decrease quality, what about xformers, with it also can get quality loss?
DI get almost the same speeds with xformers and a little more vram usage 20gb or so
Fyes in past both degraded
Fbut it is getting better
Dok thanks!! I will check if I can see any difference
Fi will do a training with onetrainer
Flets see difference between kohya
Dthat will be an interesting test!! a good test can be also to do a run test with xformers and spda, its a little speed up but can be very significant speed up for very large datasets and very long trainings.
Fdoing a fresh install
Fof onetrainer
Fye for massive dataset
Fyou have to have xformers and batch size bigger
RHi just checking to see if this works now ? And if so, how do I do it?
DLR Scheduler sdpa never worked on kohya_ss an sdxl models, seems to be an interesting concept I read that it automatically adjusts the learning rate or something like that.
DI don't know if it was designed to work in conjunction with CrossAttention sdpa
Fi use static LR
DYes I know, I think is the best choice for a little dataset and for training just one subject, but what do you think for other cases like bigger and probably unbalanced datasets with multiple subjects? a smart lr could be nice.
NI have trained a lora model in kohya and I have obtained better likeness and consistency than with a full dreambooth model
NLora is a underrated model
Fsmart LR not working for stable diffusion 
Fthey work for LLMs
Fprobably your config was not good for dreambooth. lora is scientifically proven to be inferior
NI used your presets 
FWhy I don't research LoRA training? Because if you need LoRA, do a full DreamBooth / Fine Tuning and extract LoRA. You will get better quality. This is for…
FYouTubeSECourses
In this tutorial, I am going to show you how to install OneTrainer from scratch on your computer and do a Stable Diffusion SDXL (Full Fine-Tuning 10.3 GB VRAM) and SD 1.5 (Full Fine-Tuning 7GB VRAM) based models training on your computer and also do the same training on a very cheap cloud machine from MassedCompute if you don't have such compute...
NI did too, I watched and trained with onetrainer
Nsuper weird
Dohh ok I didn't know, I'm not a programmer
NUse portrait photos taken with a sony camera
NI used my photography lights to take the photos and stuff
NBut it didn't give me a good enough likenss
NI'm doing a comparison, I'll send them later too see
DWith the right parameters and number of steps full checkpoint is better, but depending your dataset number of images, etc may be it needs more steps or something
NI know, I used just kind 16 portrait photos, also realistic vision model
DI'm not an expert but I'm talking from my own experience, I had very good results with lora too I'm not gonna lie but in my opinion Dr. Furkan Gözükara is right in this one