Software Engineering Courses (SECourses)•15mo ago

ye i think e4m3fn better https://imgsli.com/MzA3OTY0/0/2

ye i think e4m3fn better
https://imgsli.com/MzA3OTY0/0/2

Furkan Gözükara SECoursesOP•10/15/24, 8:44 PM

32 GB RAM sucks so bad

FFurkan Gözükara SECourses it was always

fadounav•10/15/24, 8:53 PM

ok because you told me that (For stylized formations the flow is still not great) thank you very much

FFurkan Gözükara SECourses ye i think e4m3fn better https://imgsli.com/MzA3OTY0/0/2

fadounav•10/15/24, 8:54 PM

That's a good stuff to compare

Ffadounav ok because you told me that (For stylized formations the flow is still not great...

Furkan Gözükara SECoursesOP•10/15/24, 8:54 PM

well if you want fully stylized person generation

Furkan Gözükara SECoursesOP•10/15/24, 8:54 PM

you should train on stylized images

Furkan Gözükara SECoursesOP•10/15/24, 8:55 PM

first do fine tuning

Furkan Gözükara SECoursesOP•10/15/24, 8:55 PM

generate stylized images

Furkan Gözükara SECoursesOP•10/15/24, 8:55 PM

retrain

Furkan Gözükara SECoursesOP•10/15/24, 8:55 PM

by the way fine tuning yields way better stylization

Furkan Gözükara SECoursesOP•10/15/24, 8:55 PM

than lora

fadounav•10/15/24, 8:57 PM

@Dr. Furkan Gözükara Yes, I'm at that stage, I've trained several stylized characters, but I would have liked to train the style in order to create object environments and the like. The aim is to create a world in the same style on Comfy UI.

fadounav•10/15/24, 8:59 PM

Like this

Ffadounav Like this

Furkan Gözükara SECoursesOP•10/15/24, 9:05 PM

so if you can collect such images as dataset

Furkan Gözükara SECoursesOP•10/15/24, 9:05 PM

it will work perfect

fadounav•10/15/24, 9:12 PM

@Dr. Furkan Gözükara Ok here's my theory, I train a Dreambooth with a character I call X for example, then I train this model with another character I call Y, always in the same style. With the captions well detailed, the goal is to have a final model that would allow me to make X and or Y appear via the prompt, is this consistent? or is it better to go through lora? btw I can have 256 images per character, if I win against my laziness lol

Ffadounav @Dr. Furkan Gözükara Ok here's my theory, I train a Dreambooth with a character ...

Furkan Gözükara SECoursesOP•10/15/24, 9:16 PM

for SDXL it can work

Furkan Gözükara SECoursesOP•10/15/24, 9:16 PM

for FLUX they bleed

Furkan Gözükara SECoursesOP•10/15/24, 9:16 PM

also you may train all at once

Furkan Gözükara SECoursesOP•10/15/24, 9:16 PM

assign differnt unique toknes

Furkan Gözükara SECoursesOP•10/15/24, 9:16 PM

ohwx man

Furkan Gözükara SECoursesOP•10/15/24, 9:16 PM

bbuk man

Dazzastrous [4090]•10/15/24, 9:17 PM

just merge sdxl and flux together lol Fluxl

DDazzastrous [4090]just merge sdxl and flux together lol Fluxl

fadounav•10/15/24, 9:18 PM

Haha xcellent

FFurkan Gözükara SECourses assign differnt unique toknes

fadounav•10/15/24, 9:21 PM

interesting, I could note the token for each character in the captions, but then I'd also have to note it separately in the training parameters? Is there a risk of it bleeding?

Ffadounav interesting, I could note the token for each character in the captions, but then...

Furkan Gözükara SECoursesOP•10/15/24, 9:27 PM

risk of bleeding exists

Furkan Gözükara SECoursesOP•10/15/24, 9:27 PM

perhaps

Furkan Gözükara SECoursesOP•10/15/24, 9:27 PM

ohwx man

Furkan Gözükara SECoursesOP•10/15/24, 9:27 PM

bbuk woman

Furkan Gözükara SECoursesOP•10/15/24, 9:27 PM

bbuk person

Furkan Gözükara SECoursesOP•10/15/24, 9:27 PM