I am pretty sure I did a full Finetune with 1x A6000 the other day😅
I am pretty sure I did a full Finetune with 1x A6000 the other day
EAnd the model came out pretty good, full 23GB checkpoints
F1x uses 27.5 gb
F2x uses 77.5
Fbecause optimizations not working
Fi am talking with kohya he couldnt solve yet
EOh alright I see, so just 1 A6000 gpu is working for dreambooth currently
Fit works even for 6 gb gpus :d
Fi made a very detailed publish just few minutes ago
SHow much does a full fine tune cost on mass compute/runpod?
Fi think read it entirely
Fpublished it
SOh thanks
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
EI think I paid around 3 - 4$
Fthis is a gold post took me more than 1 week
Fas low as 1 $ depends on dataset
SMeans that I have to be a gold member to see the post? Sorry I am entirely new to Patreon
SOh, I see, I can see the text but I can't access the links
EYou can access it even with lowest supporter tier
Fno all members can see and access
SWhen I do a fine tune, I can download the checkpoint and use it in comfyui, right?
SBut the checkpoint size is around the same size as flux?
MHello community, I am having a 2gb lora on SD , it was like 147mo only for 40mn ,
what I am doing wrong now ? the file is bigger and takes about 3h , the lora does not even look alike
heeeelp
what I am doing wrong now ? the file is bigger and takes about 3h , the lora does not even look alike
heeeelp
Fyes
F23.8 gb
Flora network dimension
Fincreases size
MI used dreambooth training ( not lora only ) Tab 1 not 2 , what should I use ( for face only lora )
Mit worked you are the best @Dr. Furkan Gözükara
D@Dr. Furkan Gözükara Update on my https://huggingface.co/nyanko7/flux-dev-de-distill training tests. Some bad news and some good news, Good news: I can train many subjects at the same time without bleeding between each other 11 people in one LORA, simple captions "Name class" note: the token must be different a saw a little bleeding with similar names like "Diego man" "Dani man" works best with "NameLastname class" so they end up been very different. Bad news: there is still some class bleeding, may be my fault because I was using a higher lr than recommended to get faster results, Other thing I was using faster presets rather than quality presets "Apply T5 Attention Mask disabled" "Train T5-XXL disabled" now I'm testing with dose enabled, other bad news regularization images still reduces resemblance, I will try this again with "Apply T5 Attention Mask enabled" and "Train T5-XXL enabled" and with the recommended learning rate. The other option that I didn't try was "Guidance scale for Flux1" flux-dev-de-distill has a default cfg of 3.5, that option I leave it at 1 for now because I'm using regular Flux-Dev for inference is another thing I have to test. I will report my Updates.
Fhopefully it will be my next research after i make tutorial
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
DSo you add the image captions, is that correct? Literally "NameLastname class"? so "JohnDoe class"?
Dyes thats it
D"NameLastname man" "NameLastname woman" etc.
SI have questions about Flux Lora training especially about prompting for characters:
SDo you prompt at all?
Do you use tags or natural language?
Do you use tags or natural language?
SWhat do you describe in the prompt?
Everything that is important about the character or everything that is not important about the character?
Or everything in the training picture?
Everything that is important about the character or everything that is not important about the character?
Or everything in the training picture?
DI use "Charactername class and the normal prompt" or "Charactername class, prompt"
Din the prompt put all you want, the clothing the setting location what is doing, etc
Sare you using kohya? if so how do you feel these boxes:
Insance prompt: diego man,
and class prmpt: man .
Insance prompt: diego man,
and class prmpt: man .
DInsance prompt: Diego
Dclass prmpt: man
Dthat's for training
Dthen promt like I said
DDr. Furkan have a great list of prompts
FHey, @Dr. Furkan Gözükara I think 256 images is a lot, can I make 64 for a Dreamooth?