Software Engineering Courses (SECourses)•2y ago

i feel like the resolution of the software got altered or something edit : ok my chrome decided to p

i feel like the resolution of the software got altered or something
edit : ok my chrome decided to put comfy ui in 15% scale resolution, super weird, but it fixed it to put it back to 100%

TThéo i feel like the resolution of the software got altered or something edit : ok my...

Furkan Gözükara SECourses•7/27/24, 12:32 AM

i was gonan tell this

FFurkan Gözükara SECourses loss rates were never meanginful for me

dsienra•7/27/24, 12:37 AM

yes may be, but I was able to see some correlation on loss graphs, they tend to get lower over time during training, on onetriner and sd3 the graph looks ridiculous, is a mess goes up and down and the model gets destroyed very quickly

dsienra•7/27/24, 12:39 AM

kohya_ss on the sd3 branch is very usable, it need some tweaks but is getting there

Ddsienra kohya_ss on the sd3 branch is very usable, it need some tweaks but is getting th...

Furkan Gözükara SECourses•7/27/24, 12:39 AM

i am still waiting him to fix

Furkan Gözükara SECourses•7/27/24, 12:50 AM

Fabricatedgirls•7/27/24, 5:28 AM

Hey Dr. Furkan do you have a good article on captioning/training photographic styles in SDXL?

Fabricatedgirls•7/27/24, 5:28 AM

like for types of lenses or lighting, stuff like that

Ddsienra I'm testing a SD3 full finetune with included kohya_ss SD3 preset, and as you ca...

protector131090•7/27/24, 7:26 AM

just change the res. i trained it many times. quality is super good. and it trains 3x times faster than xl.

FFabricatedgirls Hey Dr. Furkan do you have a good article on captioning/training photographic st...

Furkan Gözükara SECourses•7/27/24, 7:56 AM

not yet sadly

Furkan Gözükara SECourses•7/27/24, 8:03 AM

automatic1111 new version published : https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.10.0

GitHub

Release 1.10.0 · AUTOMATIC1111/stable-diffusion-webui

Features:

A lot of performance improvements (see below in Performance section)
Stable Diffusion 3 support (#16030, #16164, #16212)

Recommended Euler sampler; DDIM and other timestamp samplers cur...

Pprotector131090 just change the res. i trained it many times. quality is super good. and it trai...

dsienra•7/27/24, 12:01 PM

great!! thanks. I'm just finishing a test at 1024x1024

Marc-Daniel Schulz•7/27/24, 9:39 PM

Hey guys, I have a question for you. When training a SD Model and the results turn out somewhat disformed. What could that be?

Marc-Daniel Schulz•7/27/24, 9:39 PM

Marc-Daniel Schulz•7/27/24, 9:40 PM

MMarc-Daniel Schulz Hey guys, I have a question for you. When training a SD Model and the results t...

Furkan Gözükara SECourses•7/27/24, 10:02 PM

resolution mismatch

Furkan Gözükara SECourses•7/27/24, 10:03 PM

what do you train sdxl sd 1.5? whichn model

Marc-Daniel Schulz•7/27/24, 10:03 PM

XL Pony realism

Marc-Daniel Schulz•7/27/24, 10:03 PM

trained with 1024

Marc-Daniel Schulz•7/27/24, 10:03 PM

rendered 1280:768

Marc-Daniel Schulz•7/27/24, 10:03 PM

hey and nice talking to you. mr stablediffusionsuperman. thanks for your work

MMarc-Daniel Schulz XL Pony realism

Furkan Gözükara SECourses•7/27/24, 10:09 PM

pony must be the reason

MMarc-Daniel Schulz hey and nice talking to you. mr stablediffusionsuperman. thanks for your work ❤️

Furkan Gözükara SECourses•7/27/24, 10:09 PM

thanks

Furkan Gözükara SECourses•7/27/24, 10:09 PM

did you try 1024x1024 render?

Furkan Gözükara SECourses•7/27/24, 10:09 PM

same issue?

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

real vis xl4 is very good for training a person

Marc-Daniel Schulz•7/27/24, 10:10 PM

aight. I will try that

Marc-Daniel Schulz•7/27/24, 10:12 PM

i was wondering if it have to do with learningrate or bad captioning.

Marc-Daniel Schulz•7/27/24, 10:12 PM

but

Marc-Daniel Schulz•7/27/24, 10:12 PM

same settings but 1024.

Marc-Daniel Schulz•7/27/24, 10:12 PM

but still deformed

Marc-Daniel Schulz•7/27/24, 10:12 PM

ill give realvis xl4 a shot :=

Marc-Daniel Schulz•7/27/24, 10:12 PM

MMarc-Daniel Schulz 🙂

Furkan Gözükara SECourses•7/27/24, 10:30 PM

ye try it

Furkan Gözükara SECourses•7/27/24, 10:30 PM

if still same problem

Furkan Gözükara SECourses•7/27/24, 10:30 PM

it could be related to dataset or training setup error

Marc-Daniel Schulz•7/27/24, 10:31 PM

ill give it another shot and report back

Marc-Daniel Schulz•7/27/24, 10:31 PM

thanks

MMarc-Daniel Schulz ill give it another shot and report back

Furkan Gözükara SECourses•7/27/24, 10:36 PM

sure

jvmbo•7/28/24, 1:08 AM

Is there any quality comparison between “roop” and “rope”?

Jjvmbo Is there any quality comparison between “roop” and “rope”?

Furkan Gözükara SECourses•7/28/24, 1:25 AM

i dont have any yet but you can 1 click install test. if you do let us know

Maxivi•7/28/24, 5:26 AM

@Dr. Furkan Gözükara do you still find using no xformers is still better\

Maxivi•7/28/24, 5:26 AM

conducted another 5 tests seems xformers is still not good

MMarc-Daniel Schulz but

Maxivi•7/28/24, 5:27 AM

are your base input images all 1024x1024 or do you have different resolution images

FFurkan Gözükara SECourses try juggernaught export

__pirate_king__•7/28/24, 3:57 PM

Could you please elaborate? How exactly could I try using it?

MMaxivi @Dr. Furkan Gözükara do you still find using no xformers is still better\

Furkan Gözükara SECourses•7/28/24, 4:06 PM

i didnt test

Furkan Gözükara SECourses•7/28/24, 4:06 PM

but it is likely to reduce quality

___pirate_king__Could you please elaborate? How exactly could I try using it?

Furkan Gözükara SECourses•7/28/24, 4:06 PM

in this tutorial you will see what i mean

Furkan Gözükara SECourses•7/28/24, 4:06 PM

https://youtu.be/rjXsJ24kQQg

YouTubeSECourses

Instantly Transfer Face By Using IP-Adapter-FaceID: Full Tutorial &...

Welcome to the comprehensive tutorial on IP Adapter Face ID!

In this detailed video, I unveil the secrets of installing and utilizing the experimental IP Adapter Face ID model. This model uniquely integrates ID embedding from face recognition, replacing the conventional CLIP image embedding. Whether you're a beginner or a pro, this guide is ...

Furkan Gözükara SECourses•7/28/24, 4:06 PM

28:26 How to convert a custom CivitAI or any model into Diffusers on Kaggle to use

Furkan Gözükara SECourses•7/28/24, 4:06 PM

20:58 How to convert CKPT or Safetensors checkpoints into diffusers on RunPod