Is it a bad idea to use regularisation dataset with a cartoon dataset for fine tuning?
Is it a bad idea to use regularisation dataset with a cartoon dataset for fine tuning?
Fit depends. for flux we dont use. for sdxl if your reg images are also cartoon good
LI admire your work. As a Patreon member and Massed Compute user ($100+ spent), I'm experiencing an issue with my Dreambooth training.
My training with 277 properly prepared images, using background filling, is progressing using the "Fine Tuning / DreamBooth: 256 Training Images & Batch Size is 1 : Best Epoch is 70" setting. This should finish within the hour on Massed Compute, totaling 277 x 70 steps. I have checkpoint intervals every 10 epochs and have six models so far.
The first checkpoint showed excellent results, but subsequent epochs have produced increasingly blurry images.
Would it be possible to discuss this with you, perhaps via Skype? I would appreciate any guidance you can offer, either freely or through a specialized support option.
My training with 277 properly prepared images, using background filling, is progressing using the "Fine Tuning / DreamBooth: 256 Training Images & Batch Size is 1 : Best Epoch is 70" setting. This should finish within the hour on Massed Compute, totaling 277 x 70 steps. I have checkpoint intervals every 10 epochs and have six models so far.
The first checkpoint showed excellent results, but subsequent epochs have produced increasingly blurry images.
Would it be possible to discuss this with you, perhaps via Skype? I would appreciate any guidance you can offer, either freely or through a specialized support option.
Fthank you so much and welcome. replied you from email
�can you explain this, why are the eyes so bad? did sdxl lora training, 30 train pics, all great quality, many close ups on face, 150 epochs, repeat 1, 1024 resolution, biglovexl2 as base model.
�training on civitai
�training-data, i see network rank 32. don't know if its possible to change that on training
train-data.txt1.41KB
Fare you using our config_?
Fwhich config name?
Falso some people reporting bad results
Fi am gonna train now to see
�this was civitai training, I used what you said about ohwx woman token, 150 epochs, repeat 1.
�where can we download flux dev fast now? black labs has it gated, so can't use the download link in script...
Fah i dont know
Fi trained on my local pc yesterday
Four config
Flatest kohya
Fperfect results
�Oh awesome, you recommend kohya for sdxl lora and checkpoint training now?
Fi tested flux
Fbut we have sdxl dreambooth config too
�Ok, so onetrainer gives slightly better results for sdxlin your opinion?
AHey Doc, how you doing today, I was just wondering if it’s possible to create a stylised flux model in place of a stylised lora
LI am pleased to inform you that the dreambooth training has produced excellent results. It seems the previous difficulty arose from an incorrect learning rate.
I have encountered a remaining challenge: all generated images tend to resemble the model, irrespective of the prompt utilized. Although the model was trained using the prompt "ohwx" and the class "woman," this issue persists even when the prompt is omitted or when a different subject, such as "man," is specified.
Could you please offer some guidance on the potential cause of this occurrence? My objective is to generate images of other women/man in certain scenes who do not precisely replicate the trained model.
I have encountered a remaining challenge: all generated images tend to resemble the model, irrespective of the prompt utilized. Although the model was trained using the prompt "ohwx" and the class "woman," this issue persists even when the prompt is omitted or when a different subject, such as "man," is specified.
Could you please offer some guidance on the potential cause of this occurrence? My objective is to generate images of other women/man in certain scenes who do not precisely replicate the trained model.
Fsurely possible
Fyes it will make everyone same
Fthat is biggest issue of flux
Fit bleeds horribly
1Have you had a chance to try hidream yet? I’ve been testing loras with the same data set as flux and it’s so much better. Not sure how it will cope with multiple subjects concepts but for my purpose it’s miles ahead.
Fi didnt train yet
Fplanning to research it deeply
Fcurrently flux is just perfect except multi subject
1I think hidream is going to pull ahead once people start tuning it
Fi am adding hidream to my downloader now :d
Fhopefully will make a tutorial today
Ffor how to download use swarmui downloader
�do I still need regularization images when training checkpoint/lora for a woman using sdxl checkpoint?
JAnyone else suddenly having issues fine tuning with kohya on Massed compute? I've done this like 50 times now the past 3 months and am suddenly getting errors
Janyone have any idea what I might be doing wrong?
JLooks like somethings up with the new repo updates to the models
�Thank you Dr. And grabbing the 1024px ones will be sufficient. No point going higher. And you already cropped to make the person occupy most of the photos yes? So I don't have to run your ultimate image process on them.
�doing my first sdxl fine tune with using regimages. i had 54 training images, so it looks like it's using 55 reg images per epoch, does this look right? don't wanna waste 25h 
��Time changed to 6h. Nice
Fjust fixed today
F54 + 55 = 99
Fyou do only 10 extra :d