Software Engineering Courses (SECourses)•15mo ago

I think its torch 2.5 it works on 2.4 - but slow. increase block swap to 36 didn't help. But that w

I think its torch 2.5 it works on 2.4 - but slow. increase block swap to 36 didn't help. But that was the culprit 1024 vs 768. Must just be slightly short vram. Or have something to do with source images being PNG? is your set jpg?

TtheJindesan I think its torch 2.5 it works on 2.4 - but slow. increase block swap to 36 did...

Furkan Gözükara SECourses•10/26/24, 11:49 PM

no image doesnt matter

Furkan Gözükara SECourses•10/26/24, 11:49 PM

torch 2.4 works but way slower

Furkan Gözükara SECourses•10/26/24, 11:49 PM

so maybe that is why

FFurkan Gözükara SECourses no image doesnt matter

theJindesanOP•10/26/24, 11:52 PM

I am trying with a smaller source image set to see if makes a difference. 15 vs 30 images.

TtheJindesan I am trying with a smaller source image set to see if makes a difference. 15 vs...

theJindesanOP•10/27/24, 12:41 AM

so even with 5 images 1024/1024 fails, but 768/768 works with any number of images on 8GB torch 2.5 . I even changed OS to windows 11 to try to figure out.. maybe its a memory leak in torch 2.5 and older nvidia.

TtheJindesan so even with 5 images 1024/1024 fails, but 768/768 works with any number of imag...

Furkan Gözükara SECourses•10/27/24, 12:41 AM

very possible

Furkan Gözükara SECourses•10/27/24, 12:41 AM

i got to sleep now

theJindesanOP•10/27/24, 12:46 AM

FYI precompiled pytorch - does not require cuda installed on windows. it comes built in.. I have completely removed VS2022 and cuda from my system during troubleshooting

TtheJindesan FYI precompiled pytorch - does not require cuda installed on windows. it comes ...

Furkan Gözükara SECourses•10/27/24, 9:49 AM

true

Furkan Gözükara SECourses•10/27/24, 9:49 AM

cuda is used only when you compile something

Furkan Gözükara SECourses•10/27/24, 9:49 AM

like insightface

Furkan Gözükara SECourses•10/27/24, 9:49 AM

xpose

Furkan Gözükara SECourses•10/27/24, 9:49 AM

etc

Xcorpion•10/27/24, 6:37 PM

4060 ti (16gb) vs 4070 (12gb)
which is faster when we consider about lora traning and stuff, I mean like just having a big vram means faster train?

XXcorpion 4060 ti (16gb) vs 4070 (12gb) which is faster when we consider about lora tranin...

Furkan Gözükara SECourses•10/27/24, 6:37 PM

16 gb

Furkan Gözükara SECourses•10/27/24, 6:38 PM

if VRAM is not enough you have to sacrifice a lot of speed

Xcorpion•10/27/24, 6:38 PM

I see

Xcorpion•10/27/24, 6:41 PM

oh and was wondering if we got 2gpus, can we like combine or do smthing in training?

XXcorpion oh and was wondering if we got 2gpus, can we like combine or do smthing in train...

Furkan Gözükara SECourses•10/27/24, 6:43 PM

nope sadly you cant

Furkan Gözükara SECourses•10/27/24, 6:43 PM

you can maybe multi GPU SDXL train

Furkan Gözükara SECourses•10/27/24, 6:43 PM

or SD 1.5

Furkan Gözükara SECourses•10/27/24, 6:43 PM

with 2x 16gb GPUs

Furkan Gözükara SECourses•10/27/24, 6:43 PM

but it may not bring any speed

Furkan Gözükara SECourses•10/27/24, 6:43 PM

on windows

Xcorpion•10/27/24, 6:54 PM

ah sad...

dawnslayerz•10/27/24, 7:10 PM

i keep trying to install kohya with flux per instructins but after i copy paste the first step, it ends with this error "Copying accelerate config file to: /root/.cache/huggingface/accelerate/default_config.yaml
Traceback (most recent call last):
File "/workspace/kohya_ss/kohya_gui.py", line 6, in <module>
import gradio as gr
ModuleNotFoundError: No module named 'gradio'
root@c9368cea187b:/workspace# "

Ddawnslayerz i keep trying to install kohya with flux per instructins but after i copy paste ...

Furkan Gözükara SECourses•10/27/24, 7:14 PM

get a fresh pod

Furkan Gözükara SECourses•10/27/24, 7:14 PM

reinstall

Furkan Gözükara SECourses•10/27/24, 7:15 PM

he just fixed errors

Saxon•10/27/24, 7:43 PM

I'm trying to train a lora/dreambooth to make simple icon type images of a particular style, and having watched the tutorials and looked at the sample outputs, I'm not sure if this project is better suited to dreambooth or lora training. The attached image is an example of one of the training images.

dawnslayerz•10/27/24, 7:54 PM

@Furkan Gözükara SECourses what template do i choose

SSaxon I'm trying to train a lora/dreambooth to make simple icon type images of a parti...

Furkan Gözükara SECourses•10/27/24, 9:45 PM

fine tuning better for everything

Furkan Gözükara SECourses•10/27/24, 9:45 PM

you can do 100%

Ddawnslayerz <@205854764540362752> what template do i choose

Furkan Gözükara SECourses•10/27/24, 9:45 PM

for which one?

FFurkan Gözükara SECourses you can do 100%

Saxon•10/27/24, 9:47 PM

I keep seeing people saying that it's best to do no captioning when training a style, but it seems odd to me -- would I not want to give it at least a minimal caption, for example "[triggerword], cow, bat wings"

Dazzastrous [4090]•10/27/24, 9:48 PM

"with pig snout"

SSaxon I keep seeing people saying that it's best to do no captioning when training a s...

Furkan Gözükara SECourses•10/27/24, 9:49 PM

flux has internal captioning mechanism

Dazzastrous [4090]•10/27/24, 9:49 PM

get to bed doc lol

DDazzastrous [4090]get to bed doc lol

Furkan Gözükara SECourses•10/27/24, 9:49 PM

still got some time :d

Furkan Gözükara SECourses•10/27/24, 9:49 PM

i need to start few trainings

Dazzastrous [4090]•10/27/24, 9:50 PM

do u have blue light protection in your lenses ?

Saxon•10/27/24, 9:53 PM

The only way I've gotten it to work close to the way I want it to is using the flux gym on default settings, but I would prefer to use Kohya -- based on the preferred results from flux gym, should I increase the number of repeats (from 1) and decrease the number of epochs (from 200)