200epoch and 4000 and you should be good timson
200epoch and 4000 and you should be good timson
T4000 what?
TI mean it's kinda working but the result is much much worse than flux dev for me in both character similarity and model creativity
Asteps
Aidk guys im getting good results
Sme too actually, krea is my new go-to for finetunings
CI did so far just Lora. Not trained yet full model dreambooth
�ok i'll do krea next for comparison. doing my first fine tune locally, done loads on massedcompute. got 5090 and 128gb ram. went with the 24gb config. 25 pics, 200 epochs looks like its going to be around 6h.
Shey @Dr. Furkan Gözükara, could you give me some advice i have a big checkpoint to train (220 epochs with 300+ images) and it would take 88hours with a L40S. Is there a gpu I can rent what would significantly speed up the training (without going multi gpu, i read here that it impacts the quality of the training) ? Thanks
Fwell 220 epoch is too big for 300 images
Fit is a single concept right?
Fthere isnt much you can do to speed up even batch size wont help much
Funless multiple gpu
S@Dr. Furkan Gözükara It's for an editorial style model so the dataset isn't very homogenous... I went with 220 because i'm training for flux krea
Swould multiple gpu hurt the quality of the training ?
Fyou can do this
F4 gpus
Fbatch size 1
Fit will minimally impact
Fso your total batch size will be 4
Fmake learning rate = learning rate * 2
SOk @Dr. Furkan Gözükara thanks a lot. Is there a patreon post where you explain how to set up multiple gpu for runpod ?
FYouTubeSECourses
Unlock the power of FLUX LoRA training, even if you're short on GPUs or looking to boost speed and scale! This comprehensive guide takes you from novice to expert, showing you how to use Kohya GUI for creating top-notch FLUX LoRAs in the cloud. We'll cover everything: maximizing quality, optimizing speed, and finding the best deals. With our exc...
Syou're the best
SJust to make sure, I was using 4e-06, what would that make x2 ?
F4e-06 is 2x of 2e-06
STHanks 
S@Dr. Furkan Gözükara there are no 4xGPU config file in the latest kohya zip file
Fnot for dreambooth
Fit needs 80gb+
Flora has it
F
Fbut you can follow logic of lora and test
Sok will do thanks
�my first image was ok, the 2nd one looked like this, why?
�hmm, maybe 16bit enabled with the t5xxx_enconly caused it?
�or this happens, normal image fine, yolo face segment gets pixelated. whats wrong
�first pic ok, except for face, next pic with exact same settings and prompt, not even normal pic is fine
Ayoure doing something wrong
�obviously. been fine-tuning since september last year and i always use swarmui for grid-comparisons. first time i see this problem. does it have to do with flux krea? am i supposed to use different clips or smth?
Awhat are you settings
�same as flux dev
Aput guidance at 4.5
�Prompt: ohwx woman in a 1950s-inspired midi dress with full skirt, small polka dots, three-quarter sleeves, peter pan collar, matching belt at natural waist, kitten heel pumps, delicate bracelet watch, pearl necklace, garden party setting with blooming roses, soft golden hour lighting<segment:yolo-face_yolov9c.pt-1,0.7,0.5>photograph of ohwx woman⧉,
Model: goth-fluxkrea-1-000100, Seed: 1991094289, Steps: 40, CFG Scale: 1, Aspect Ratio: Custom, Width: 1280, Height: 1600, Side Length: 1280, Sampler: iPNDM (Improved Pseudo-Numerical methods for Diffusion Models), Scheduler: DDIM Uniform, Flux Guidance Scale: 4, VAE: ae, CLIP-L Model: clip_l, T5-XXL Model: t5xxl_fp16, Preferred DType: Default (16 bit)
Model: goth-fluxkrea-1-000100, Seed: 1991094289, Steps: 40, CFG Scale: 1, Aspect Ratio: Custom, Width: 1280, Height: 1600, Side Length: 1280, Sampler: iPNDM (Improved Pseudo-Numerical methods for Diffusion Models), Scheduler: DDIM Uniform, Flux Guidance Scale: 4, VAE: ae, CLIP-L Model: clip_l, T5-XXL Model: t5xxl_fp16, Preferred DType: Default (16 bit)
Aalso disable scheduler
�this was the first pic i generated, and it worked, then next pic got pixelated...
Asee guys , krea is unbeatable
Alook at that realism
SI set up 4xL40S but when starting the starting i got this for each of them :
torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 22.17 GiB. GPU 0 has a total capacity of 44.40 GiB of which 21.58 GiB is free. Process 271586 has 22.81 GiB memory in use. Of the allocated memory 22.18 GiB is allocated by PyTorch, and 5.70 MiB is reserved by PyTorch but unallocated.
I used the 48GB_GPUz_4x_GPU_Quality_Tier_1_42000MB_10.6_Second_IT_Trains_T5_and_T5_Attention file, then tried with the Quality Tier 3 and still having the same issue. Any clue ?
torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 22.17 GiB. GPU 0 has a total capacity of 44.40 GiB of which 21.58 GiB is free. Process 271586 has 22.81 GiB memory in use. Of the allocated memory 22.18 GiB is allocated by PyTorch, and 5.70 MiB is reserved by PyTorch but unallocated.
I used the 48GB_GPUz_4x_GPU_Quality_Tier_1_42000MB_10.6_Second_IT_Trains_T5_and_T5_Attention file, then tried with the Quality Tier 3 and still having the same issue. Any clue ?