Software Engineering Courses (SECourses)•15mo ago

hello i wanted to ask him a kohya related question is this bmaltais's account https://x.com/bmaltais

hello i wanted to ask him a kohya related question is this bmaltais's account https://x.com/bmaltais (dude's every post is about elon musk, spacex or tesla)

bmaltais (@bmaltais) / X

Twitter

magicalOP•10/16/24, 10:55 AM

oh it says bmaltais can't be messaged

deepblue_1337•10/16/24, 11:39 AM

Anybody has any idea why I get that error:
Traceback (most recent call last):
File "/home/Ubuntu/apps/kohya_ss/venv/bin/accelerate", line 8, in <module>
sys.exit(main())
File "/home/Ubuntu/apps/kohya_ss/venv/lib/python3.10/site-packages/accelerate/commands/accelerate_cli.py", line 48, in main
args.func(args)
File "/home/Ubuntu/apps/kohya_ss/venv/lib/python3.10/site-packages/accelerate/commands/launch.py", line 1106, in launch_command
simple_launcher(args)
File "/home/Ubuntu/apps/kohya_ss/venv/lib/python3.10/site-packages/accelerate/commands/launch.py", line 704, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/home/Ubuntu/apps/kohya_ss/venv/bin/python', '/home/Ubuntu/apps/kohya_ss/sd-scripts/flux_train.py', '--config_file', '/home/Ubuntu/Downloads/ai/training_img/model/config_dreambooth-20241016-112439.toml']' died with <Signals.SIGKILL: 9>.

snowmonkey•10/16/24, 11:40 AM

@Dr. Furkan Gözükara Have you tried getting clear text out with Flux using ControlNet ?

Something like this:
https://www.reddit.com/r/StableDiffusion/comments/119j8qr/clear_text_using_controlnet/

But with Flux

From the StableDiffusion community on Reddit: Clear text using Cont...

Explore this post and more from the StableDiffusion community

Ssnowmonkey @Dr. Furkan Gözükara Have you tried getting clear text out with Flux using Contr...

druhl•10/16/24, 12:20 PM

You can do this by direct prompting the T5

druhl•10/16/24, 12:20 PM

No controlnets required.

CClint Hardwood I gave it a go and I got this error:

Furkan Gözükara SECourses•10/16/24, 12:25 PM

you are missing git very likely

Furkan Gözükara SECourses•10/16/24, 12:25 PM

you need to follow this video for requirements

Furkan Gözükara SECourses•10/16/24, 12:26 PM

you didnt switch to accurate branch

Furkan Gözükara SECourses•10/16/24, 12:26 PM

my installer does that

Furkan Gözükara SECourses•10/16/24, 12:26 PM

https://youtu.be/DrhUHnYfwC0

YouTubeSECourses

How to Install Python, CUDA, cuDNN, C++ Build Tools, FFMPEG & Git T...

If you are interested in using AI, generative AI applications, open source applications in your computer, than this is the most fundamental and important tutorial that you need. In this tutorial I show and explain how to properly install appropriate Python versions accurately, how to switch between different Python versions, how to install diffe...

Mmagical hello i wanted to ask him a kohya related question is this bmaltais's account ht...

Furkan Gözükara SECourses•10/16/24, 12:27 PM

he just develops the gui

Ddeepblue_1337 Anybody has any idea why I get that error: Traceback (most recent call last): ...

Furkan Gözükara SECourses•10/16/24, 12:27 PM

yes

Furkan Gözükara SECourses•10/16/24, 12:27 PM

out of RAM

Furkan Gözükara SECourses•10/16/24, 12:27 PM

did you get alt config? or spot?

Ssnowmonkey @Dr. Furkan Gözükara Have you tried getting clear text out with Flux using Contr...

Furkan Gözükara SECourses•10/16/24, 12:27 PM

not yet

FFurkan Gözükara SECourses out of RAM

deepblue_1337•10/16/24, 12:47 PM

default config, 1xA6000 (not spot)

deepblue_1337•10/16/24, 12:47 PM

config: Batch_Size_7_48GB_GPU_46250MB_28.9_second_it_Tier_1.json

deepblue_1337•10/16/24, 12:47 PM

even if I change epoch to 1 it throws the error

Ddeepblue_1337 default config, 1xA6000 (not spot)

Furkan Gözükara SECourses•10/16/24, 12:56 PM

<Signals.SIGKILL: 9>. happens with RAM

Furkan Gözükara SECourses•10/16/24, 12:57 PM

actually i tested it on multiple GPU machine

Furkan Gözükara SECourses•10/16/24, 12:57 PM

maybe batch size 7 needs more than 48 GB RAM

Furkan Gözükara SECourses•10/16/24, 12:57 PM

let me test

deepblue_1337•10/16/24, 12:58 PM

I am currently running normal one, will let me know

Furkan Gözükara SECourses•10/16/24, 12:58 PM

i am staring a machine now to test

Ddeepblue_1337 I am currently running normal one, will let me know

deepblue_1337•10/16/24, 12:59 PM

*I am currently running normal one, will let you know (sorry, typo)

FFurkan Gözükara SECourses he just develops the gui

magicalOP•10/16/24, 1:01 PM

oh so he wouldn't have knowledge about training? I have a couple of questions about Flux style training can i dm you?

deepblue_1337•10/16/24, 1:01 PM

normal (not batch size 7) works

Mmagical oh so he wouldn't have knowledge about training? I have a couple of questions ab...

Furkan Gözükara SECourses•10/16/24, 2:28 PM

i have full flux style training guide did you read?

Furkan Gözükara SECourses•10/16/24, 2:28 PM

https://huggingface.co/MonsterMMORPG/3D-Cartoon-Style-FLUX

MonsterMMORPG/3D-Cartoon-Style-FLUX · Hugging Face

Ddeepblue_1337 normal (not batch size 7) works

Furkan Gözükara SECourses•10/16/24, 2:28 PM

ye looks like 48 gb not sufficient

Furkan Gözükara SECourses•10/16/24, 2:28 PM

for batch size 7

Furkan Gözükara SECourses•10/16/24, 2:30 PM

wow i am able to open multiple thinlinc client on windows 11

Ddeepblue_1337 normal (not batch size 7) works

Furkan Gözükara SECourses•10/16/24, 2:45 PM

single A6000 batch size 7 works

Furkan Gözükara SECourses•10/16/24, 2:45 PM

fine tuning

Furkan Gözükara SECourses•10/16/24, 2:45 PM

dsienra•10/16/24, 3:14 PM

@Dr. Furkan Gözükara Update on https://huggingface.co/nyanko7/flux-dev-de-distill New training, catastrophic failure with Apply T5 Attention Mask, and Train T5-XXL enabled. the lora bleeds and is no learning the concepts, the prompt "token class" renders random things like a dog or something on regular flux-dev, on inference using flux-dev-de-distill kinda works sometimes, one image have resembles the next bleeds, all over the place. I going to disable Apply T5 Attention Mask, and Train T5-XXL, one of this options is breaking the model. all tests images have horizontal lines also.

nyanko7/flux-dev-de-distill · Hugging Face

dsienra•10/16/24, 3:20 PM

@Dr. Furkan Gözükara may be the problem wasn't the learning rate, may be is one or both of those Apply T5 Attention Mask, and Train T5-XXL, what do you think?

FFurkan Gözükara SECourses i have full flux style training guide did you read?

magicalOP•10/16/24, 3:40 PM

I did. Is there much quality difference between using Rank 3 config vs 4x_GPU_Rank_1_SLOW?

magicalOP•10/16/24, 3:41 PM

If not the only think I didn't try is training without captions

Ddsienra @Dr. Furkan Gözükara may be the problem wasn't the learning rate, may be is one ...

Furkan Gözükara SECourses•10/16/24, 3:48 PM

i think it could be all so every case has to be tested

Mmagical I did. Is there much quality difference between using Rank 3 config vs 4x_GPU_Ra...

Furkan Gözükara SECourses•10/16/24, 3:48 PM

yes small diff

FFurkan Gözükara SECourses yes small diff

magicalOP•10/16/24, 3:50 PM

oh and I didn't try training T5 XXL. If there is no text in my training images would it still improve the training?

dsienra•10/16/24, 4:06 PM

After my failure I'm not so sure about "Apply T5 Attention Mask, and Train T5-XXL enabled" you are training the T5, but in inference you load a T5 model that doesn't know your token. forge o SwarmUI load the regular T5 model that was not trained. so I'm not so sure

Ddsienra After my failure I'm not so sure about "Apply T5 Attention Mask, and Train T5-XX...

magicalOP•10/16/24, 4:18 PM

oh the t5 model should be trained as well for it to work properly?

dsienra•10/16/24, 4:18 PM

@Dr. Furkan Gözükara may be Apply T5 Attention Mask can help, but I don't understand how can Train T5-XXL helps, because on inference you load a T5 model that was not trained . forge or SwarmUI load the regular T5 model that was not trained

Mmagical oh the t5 model should be trained as well for it to work properly?

dsienra•10/16/24, 4:23 PM

Is the thing I don't understand, so really IDK now I'm training again with both options disabled, My question can apply to everyone, but in my case is different because I'm training flux-dev-de-distill so in my case is broking something. I'm having artifacts on the images with this on and also is not learning the concepts correctly

dsienra•10/16/24, 7:04 PM

@Dr. Furkan Gözükara Update on https://huggingface.co/nyanko7/flux-dev-de-distill I started new training - "T5 Attention Mask and T5-XXL both disabled" same lr 30 epochs now. I tested the lora checkpoint and all problems are fixed, is going great using regular flux-dev-fp8 for inference, I'm training tree people same class, same prompt changing just the name renders the correct subject, is undertrained but goes very well so far

nyanko7/flux-dev-de-distill · Hugging Face

ar.1•10/16/24, 9:42 PM

ar.1•10/16/24, 9:42 PM

did i do something wrong? why is there 17.6 million steps ??

hello i wanted to ask him a kohya related question is this bmaltais's account https://x.com/bmaltais

Similar Threads

Similar Threads

Similar Threads