Software Engineering Courses (SECourses)•16mo ago

2x upscale add those noise

2351-photo20of20ohwx20man20standing20on20a20high-tec-flux1-dev-1596475883.png

Furkan Gözükara SECoursesOP•9/8/24, 11:51 PM

org

FFurkan Gözükara SECourses 2x upscale add those noise

dsienra•9/9/24, 12:05 AM

it adds horizontal banding

dsienra•9/9/24, 12:05 AM

what upscale method?

Furkan Gözükara SECoursesOP•9/9/24, 12:06 AM

2041-photo20of20ohwx20man20riding20a20majestic20mus-flux1-dev-1596475354.png

Ddsienra what upscale method?

Furkan Gözükara SECoursesOP•9/9/24, 12:06 AM

2x of swarmui

dsienra•9/9/24, 12:11 AM

did you try upscaling with supir?

Ddsienra did you try upscaling with supir?

Furkan Gözükara SECoursesOP•9/9/24, 12:15 AM

i will do that

Furkan Gözükara SECoursesOP•9/9/24, 12:15 AM

saving both for comparison later

Furkan Gözükara SECoursesOP•9/9/24, 12:16 AM

Furkan Gözükara SECoursesOP•9/9/24, 12:16 AM

man all images are just so amazing

Furkan Gözükara SECoursesOP•9/9/24, 12:22 AM

0307-photo20of20ohwx20man20riding20a20majestic20eagl-flux1-dev-1596466140.png

xlads•9/9/24, 12:23 AM

WILD!

nXonXi•9/9/24, 12:30 AM

Hi i need to hunderstand
i'm using kohya_ss gui
what i experiments it's that i cant let empty the area path repo.

NnXonXi Hi i need to hunderstand i'm using kohya_ss gui what i experiments it's that i c...

nXonXi•9/9/24, 12:33 AM

No matter what I put, it creates a folder in the root of my repo.
What to do if i just want kohya to upload safetensors in my root repo ?

NnXonXi Hi i need to hunderstand i'm using kohya_ss gui what i experiments it's that i c...

Furkan Gözükara SECoursesOP•9/9/24, 12:33 AM

i couldnt make this work actually

Furkan Gözükara SECoursesOP•9/9/24, 12:33 AM

so we have a notebook to manually upload

FFurkan Gözükara SECourses so we have a notebook to manually upload

nXonXi•9/9/24, 12:34 AM

you upload manually !!!??? in one click i guess !!!???

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•9/9/24, 12:34 AM

@Dr. Furkan Gözükara where can I learn about lora blocks, block training and layer training?

NnXonXi you upload manually !!!??? in one click i guess !!!???

Furkan Gözükara SECoursesOP•9/9/24, 12:37 AM

yep

Furkan Gözükara SECoursesOP•9/9/24, 12:37 AM

shown in tutorial

�𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮 @Dr. Furkan Gözükara where can I learn about lora blocks, block training and lay...

Furkan Gözükara SECoursesOP•9/9/24, 12:38 AM

i dont know either yet

Furkan Gözükara SECoursesOP•9/9/24, 12:39 AM

making dragon real life like so hard

0038-imagine208K20DSLR20image20of20ohwx20man20ohwx-flux1-dev-1583116719.png

FFurkan Gözükara SECourses i dont know either yet

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•9/9/24, 12:43 AM

I read in some sub reddits and discords that we can train certain layers for certain features only like face, clothes, etc. Which makes it easy to use it with other loras without bleeding and resulting in lower file sizes

Furkan Gözükara SECoursesOP•9/9/24, 12:44 AM

0043-8K20DSLR20image20of20ohwx20man20wearing20eyegla-flux1-dev-1399231184.png

�𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮 I read in some sub reddits and discords that we can train certain layers for cer...

Furkan Gözükara SECoursesOP•9/9/24, 12:44 AM

yes i saw too

Furkan Gözükara SECoursesOP•9/9/24, 12:44 AM

but kohya didnt add yet

FFurkan Gözükara SECourses but kohya didnt add yet

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•9/9/24, 12:45 AM

It will be great to explore once kohya adds this

FFurkan Gözükara SECourses Click to see attachment

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•9/9/24, 12:46 AM

Where do you get these prompts?

�𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮 Where do you get these prompts?

Furkan Gözükara SECoursesOP•9/9/24, 12:47 AM

i use chatgpt and claude

FFurkan Gözükara SECourses i use chatgpt and claude

�

𝗡𝗶𝘁𝗶𝗻 𝗣𝗮𝗿𝗲𝘃𝗮•9/9/24, 12:48 AM

What's claude?

Furkan Gözükara SECoursesOP•9/9/24, 12:48 AM

claude 3.5 bbetter than chatgpt

xlads•9/9/24, 1:07 AM

which upscaling model do you use? @Dr. Furkan Gözükara

xlads•9/9/24, 1:09 AM

tried settings for upscaling from this video https://www.youtube.com/watch?v=HKX8_F1Er_w&t=746s but the generations arent looking to good

YouTubeSECourses

Zero to Hero Stable Diffusion 3 Tutorial with Amazing SwarmUI SD We...

Do not skip any part of this tutorial to master how to use Stable Diffusion 3 (SD3) with the most advanced generative AI open source APP SwarmUI. Automatic1111 SD Web UI or Fooocus are not supporting the #SD3 yet. Therefore, I am starting to make tutorials for SwarmUI as well. #StableSwarmUI is officially developed by the StabilityAI and your m...

Xxlads tried settings for upscaling from this video https://www.youtube.com/watch?v=HKX...

Furkan Gözükara SECoursesOP•9/9/24, 1:35 AM

best is supir

Furkan Gözükara SECoursesOP•9/9/24, 1:36 AM

https://youtu.be/OYxVEvDf284

YouTubeSECourses

Complete Guide to SUPIR Enhancing and Upscaling Images Like in Sci-...

In this video, I explain how to 1 click install and use the most advanced image upscaler / enhancer in the world that is both commercially and open source available. The upscaler that I am going to introduce you is open source #SUPIR and the model is free to use. SUPIR upscaler is many times better than both paid Topaz AI and Magnific AI and you...

Furkan Gözükara SECoursesOP•9/9/24, 3:13 AM

starting testing T5 training

Furkan Gözükara SECoursesOP•9/9/24, 3:18 AM

wow it added some quite VRAM

FFurkan Gözükara SECourses wow it added some quite VRAM

dsienra•9/9/24, 3:20 AM

how much? fits on 24gb cards?

Ddsienra how much? fits on 24gb cards?

Furkan Gözükara SECoursesOP•9/9/24, 3:22 AM

i didnt test 24 gb yet

Furkan Gözükara SECoursesOP•9/9/24, 3:22 AM

but in 16 bit added like 8 gb more

Furkan Gözükara SECoursesOP•9/9/24, 3:22 AM

may barely fit to 24 gb

dsienra•9/9/24, 3:24 AM

tahnks, I'll be waiting your results

Furkan Gözükara SECoursesOP•9/9/24, 3:36 AM

7 different T5 trainings started lets see tomorrow hopefully the results

dsienra•9/9/24, 4:14 AM

I heard that onetrainer is 2x faster on flux training

dsienra•9/9/24, 4:15 AM

but there is some incompatibility with forge, is diferent lora format, and to train you need flux models on diffusion format

dsienra•9/9/24, 4:16 AM

the forge incompatibility will be fixed sood, by onetrainer or by forge idk really

Ddsienra but there is some incompatibility with forge, is diferent lora format, and to t...

JohnDopamine•9/9/24, 4:51 AM

Fixed this morning: https://github.com/lllyasviel/stable-diffusion-webui-forge/commit/44eb4ea83797b2abcc2dcc3e0bfcbed77e5f7a6e

GitHub

Support T5&Clip Text Encoder LoRA from OneTrainer · lllyasviel/stab...

requested by #1727
and some cleanups/licenses
PS: LoRA request must give download URL to at least one LoRA

Buğra Alperen Uluırmak•9/9/24, 6:41 AM

Hello, I am working on fine tuning on stable diffusion xl and I am using my own interface. I have a couple of problems. I want to limit the number of colors the user wants. For example, if the user chooses 5 colors, the model should use only 5 colors can be less. The other problem is the tiling process. There are a few problems in this too. I would be happy if you help.

Ddsienra I heard that onetrainer is 2x faster on flux training

Furkan Gözükara SECoursesOP•9/9/24, 9:33 AM

just impossible with same config

FFurkan Gözükara SECourses best is supir

slipkn67•9/9/24, 9:46 AM

what is best quality in flux.dev: 1024x1024 +upscale or generate in 2048x2048 native?

2x upscale add those noise

Similar Threads