Lthank you
and if i do batch size=2, do i need to adjust learning rates? like double them
and if i do batch size=2, do i need to adjust learning rates? like double them
Ftry square root of batch size
MHi! I just downloaded Kohya and all necessary models in my Runpod. How can I use best configs json file(e.g. 48GB_GPU_Quality_Tier_1_40000MB_9.3_Second_IT_Trains_T5_and_T5_Attention) in Kohya?
Should I manually put all the parameters?
Should I manually put all the parameters?
MAnd I tried this config with L40S, but keep shows "torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 72.00 MiB.
GPU 0 has a total capacity of 44.52 GiB of which 24.25 MiB is free.
Process 165082 has 44.49 GiB memory in use."
GPU 0 has a total capacity of 44.52 GiB of which 24.25 MiB is free.
Process 165082 has 44.49 GiB memory in use."
Fyes use lesser config
Fprobably your images are not all 1024x1024
MI just checked and it's all 1024 * 1024. Is there any possibility that I'm facing OOM even though I'm using L40S(And I checked the memory usage right before the training)
Fwell it uses 40 GB so you shouldnt
Fcan you try restart machine and try again?
Fbut maybe kohya made some updates and now it uses more VRAM
MI'll try and let you know!
M@Furkan Gözükara SECourses When I change config to "cache_latents = false / cache_latents_to_disk = false", GPU usage dropped to 36GB. Is there any quality drop if I change it like this?
ZHi @Furkan Gözükara SECourses
Thank you for the dreambooth training tutorial, it really very clear and I successfully manage the training for a person and it performs really well
I have question for other case, if I want to train a person with specific clothing style and also in a specific illustration style, what will be the best approach? Shall I train first the illustration style, then the person with his specific clothing style? The aim is I want to generate this person in consistent clothing style and in consistent illustration style.
Thank you
Thank you for the dreambooth training tutorial, it really very clear and I successfully manage the training for a person and it performs really well
I have question for other case, if I want to train a person with specific clothing style and also in a specific illustration style, what will be the best approach? Shall I train first the illustration style, then the person with his specific clothing style? The aim is I want to generate this person in consistent clothing style and in consistent illustration style.
Thank you
Ffor illustration style use kontext model
Fit really does good job
Ffor clothing if you have pictures of person with that clothing
Fthat would be great to train
Fotherwise you can try to train with cloth pictures
For traing a clothing model and inpaint images of person
ZAh I see. Previously I already prepare the character with correct clothing and the desired illustration style. But I have only this one image.
I plan to create other pose and other situation using Flux Kontext. Will it perform better if I do the train in one go, it has full body clothing style and also illustration style?
I plan to create other pose and other situation using Flux Kontext. Will it perform better if I do the train in one go, it has full body clothing style and also illustration style?
MIs there best config json file for batch 7? I can only find best config for batch 1 in the folder(Kohya_FLUX_DreamBooth_LoRA_v30)
F1 in go is best
Fbut if classes are matching it can bleed
Ffor lora i recommend multiple gpu
Fto speed up
Ffor lora i dont have batch config but you can use batch size 1
Fand set batch size 7
Fand multiply learnign rate with
Fsquare root 7 * base learning rate
Fhowever batch size 1 has the best quality
SI am using your Kaggle Kohya ss file to start a kohya ss gradio ui, but it fails by saying no conda environment found, and if i manually start i and try to load a json file forf the configuration, it says something like 'str' has not attribute 'get'? please help
S
Fwhich versioın are you using?
Ffile name
Si guess its v25, i just ran it using the py file let me try
Sits 2 am now, i will let you know later what happens
Flatest is v28
Fuse it
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
MIs it even possible to set batch size 7 with VRAM 48GB? Because even with batch size 1, VRAM usage is around 36GB
MFYI, I want best quality, but I should also consider training speed
Fi think not with lora
Fit is for dreambooth at 1024x1024
MOh, I see
MFor dreambooth, batch size 1 & 7 gpu vs batch size 7 & 1 gpu
MIs this same speed?
MI assume quality would be better with batch size 1
Fof course not