Whenever i train kohyass lora on a flux model, i get error "returned non-zero exit status 1."
Whenever i train kohyass lora on a flux model, i get error "returned non-zero exit status 1."
Mother models it works perfectly fine
AI'm getting two additional options for Kohya that don't appear on the exapmle image on the guide on Patreon - "Conv Dimension" and "Minimum Difference". Should I leave them with default values?
Fyes batch size 1 is best
Fi dont know good one atm
Fthe logic is like this
Fwe do up to 200 epochs
Fand we compare and find best one
Fbut if you have too many images
Fit is becoming redundant
Fwe have image similarity app you can try
Four massed compute image has swarmui
Fjust click updat first
Fthen ready
Fgithub not found https://github.com/xh9998/DiffVSR
Fhard to know without seeing everything
Fyou want to extract flux lora?
Fthat is inaccurate tab
Fclick this
F
AOops, my bad 
. Thanks 
. Thanks 
A@Furkan Gözükara SECourses do you think to have 2 Lora people in one picture it's best to do face inpainting or regional prompting on Swarm? what will achieve best results
Fprobably face inpainting
Awith yolo segmentation or individual swarm masking using flux inpaint model?
M
M
V@Furkan Gözükara SECourses finetuning can or cannot use 2 gpu? Not used in my case. Are there any settings?
Ait can't
Jyou guys ever get this error while training? in this session im running now, seems to happen almost every time, just have to click "generate" a few times and then it works anyway?
(4x RTX 6000 on MassedCompute, using Init Image)
(4x RTX 6000 on MassedCompute, using Init Image)
PHi guys, I made this little toy for you: https://huggingface.co/spaces/elismasilva/mixture-of-diffusers-sdxl-tiling
M@Furkan Gözükara SECourses have you tried the recently released flux pro model's api for fine tuning, wondering what your take is on that vs the dreambooth method in terms of quality
Fpeople tested and it is inferior
Mi see, powering through your nearly 3 hour dreambooth tutorial on flux since yesterday haha
Fsana controlnet published
Fi should implement
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=40)
Dnice
Freplied all emails and messages time to sleep
V@Furkan Gözükara SECoursesEDIT SOLVED: In the folder on massed compute there was hidden folder "cache" that I had to delete. then I could upload again. ORIGINAL MESSEAGE: I am trying to transfer checkpoints from massed compute to huggingface. I use the jupyter notebook. Works very good and fast. But I have one problem.... Some checkpoints I had to delete from hugging face. Then I tried again to retransfer them from massed compute to hugging face. but the jupyter always says: "Recovering from metadata files: 100%". and then its just skipping.... What do I need to do? is there an overwrite command? or do i need to delete metadata files and where are they?
Mdo you think a dreambooth or lora training of just text can create a reliable text generating model?
.@Furkan Gözükara SECourses if I set the guidance scale to 6 in the kohya fine tune parameters, will that influence training and if yes how?
K
Kanyone else get python crashing when they load swarmui on the first time? running it locally
K4060ti... I think it tries to load the ckpt when i click generate image and then python crashes @Furkan Gözükara SECourses
Kmy disk that has swarmui and my .ckpts gets like 96% usage when i click "generate image"
Kim gonna try with my SSD, i think that might fix it
A@Furkan Gözükara SECourses
Manyone here having issue when training a lora for flux using kohyass?
Mcuz i have been having lots of errors and issues
C
