Software Engineering Courses (SECourses)•11mo ago

I rarely use Supir because I haven't figured out which configurations for certain styles are better,

I rarely use Supir because I haven't figured out which configurations for certain styles are better, I did mass tests with models and I didn't really get a good result(what's better for realism, better for illustration, cartoon drawings). If there are such configs I would take a look at them

when I have some free time, I'll test it out

phoenix•2/19/25, 7:11 AM

@Furkan Gözükara SECourses can I generate iamge with swarmui on the already generated checkpoint while training is going on

AiInfluence•2/19/25, 9:53 AM

@Timson we need your wisdom

NNeuromasters I rarely use Supir because I haven't figured out which configurations for certai...

Timson•2/19/25, 10:08 AM

agree

, I wish there were comprehensive guide like for flux, but I found supir guide lacking experiment data and considering how slow supir is (up to 2 minutes per image on L40S) it is hard to do research

AAiInfluence @Timson we need your wisdom

Timson•2/19/25, 10:11 AM

sorry, coudn't find the question you are referring to

AAiInfluence If I use the flux sigma model as checkpoint and load a Lora trained on flux dev ...

AiInfluence•2/19/25, 10:12 AM

@Timson

AAiInfluence If I use the flux sigma model as checkpoint and load a Lora trained on flux dev ...

Timson•2/19/25, 10:15 AM

I didn't explore Flux Sigma Vision, my impression was it is a model designed for special workflow with tiled generation, hence not compatible with most flux dev stuff

Timson•2/19/25, 10:18 AM

plus it's not a generalist model like flux dev, but model specifically trained for realistic close up portraits of men

Timson•2/19/25, 10:20 AM

that being said, this exists

https://civitai.com/images/56735138

Timson•2/19/25, 10:21 AM

16 megapixel

julius•2/19/25, 10:21 AM

Using massedcompute, how do I utilize two GPU's for fine tuning a model?

Jjulius Using massedcompute, how do I utilize two GPU's for fine tuning a model?

julius•2/19/25, 10:28 AM

ah wait found here https://youtu.be/-uhL2nW7Ddw?si=PeuMb2ddDNtDu8Se&t=1134
I assume the lora instructions are the same as for dream booth fine tuning?

only thing i worry about is the learning rate. if it's risky to calculate a new one by myself?

YouTubeSECourses

Blazing Fast & Ultra Cheap FLUX LoRA Training on Massed Compute & R...

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...

Jjulius ah wait found here https://youtu.be/-uhL2nW7Ddw?si=PeuMb2ddDNtDu8Se&t=1134 I ass...

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•2/19/25, 10:46 AM

Multi gpu dream booth training is only supported with 2 80gb gpu's

julius•2/19/25, 10:46 AM

ah shit okay

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•2/19/25, 11:04 AM

@Furkan Gözükara SECourses Should I do a dream booth for this outfit - https://drive.google.com/drive/folders/1esrUCyCq9E1gu0WaKxoMFFzP1GvBi4Tj?usp=sharing ? I trained like 15 Loras on this with different settings and with all of them I have the same problem they are not learning the buttons correctly.

Google Drive

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•2/19/25, 11:05 AM

Has anyone trained a dreambooth on clothes? If yes how were the results?

Timson•2/19/25, 11:37 AM

I don't think flux is the right tool for replication of objects with 100% accuracy especially with tiny details like buttons and not expecting creativity and fluctuations at all

Timson•2/19/25, 11:39 AM

It may be possible with huge dataset (200+ images) with very good diversity and massive overtraining

Timson•2/19/25, 11:40 AM

your dataset is 10 images with fairly poor diversity (same model, same lighting, same setting)

TTimson I don't think flux is the right tool for replication of objects with 100% accura...

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•2/19/25, 11:41 AM

I think that as well now after these countless trainings

AiInfluence•2/19/25, 2:52 PM

@Nitin Pareva you should train the pattern as a graphic instead and apply it to an outfit

AiInfluence•2/19/25, 2:52 PM

when I wanted to train a Stussy shirt i trained the stamp of the shirt

AiInfluence•2/19/25, 2:52 PM

and it worked brilliantly

AiInfluence•2/19/25, 2:53 PM

also , I don't see a reason if flux can copy complex facial structures why it wouldn't do the same for a pattern

julius•2/19/25, 4:30 PM

Anyone knows how it works with dogs? If I fine tune a flux model on a person, where in some photos he's with his dog ... can i then prompt for "ohwx man walking his dog" and then it will be the same dog?

Jjulius Anyone knows how it works with dogs? If I fine tune a flux model on a person, wh...

Timson•2/19/25, 5:35 PM

if it's less than 5% of training data you shouldn't worry.
Also you can edit|crop objects out if they interrsect your class or you worry about bleeding because of high percentage of traning data

Darth Teethius•2/19/25, 6:24 PM

How do you download the files you train on massed compute to your local pc? not seeing a way to do this in thinlinc

DDarth Teethius How do you download the files you train on massed compute to your local pc? not ...

AiInfluence•2/19/25, 6:37 PM

upload on hugging face

Guy Bertental•2/19/25, 9:05 PM

@Furkan Gözükara SECourses thank you so much for the great work that you did,
quick question - to tine tune flux model using LoRA with GTX 3070 Ti 8GB which configuration and model size do you think works best in a reasonable time?

TTimson if it's less than 5% of training data you shouldn't worry. Also you can edit|cro...

julius•2/19/25, 10:54 PM

hmm i mean I'd like to be able to generate a photo with this persons dog (not trying to avoid it). More just asking if the dog will be consistent in the final generated images if it's featured in some of the training data (say about 5% perhaps)?

DDarth Teethius How do you download the files you train on massed compute to your local pc? not ...

julius•2/19/25, 10:55 PM

Best way is to upload to Huggingface using the python script and Jupiter notebook like Furkan explains in his tutorials

AAiInfluence If I use the flux sigma model as checkpoint and load a Lora trained on flux dev ...