Software Engineering Courses (SECourses)•3y ago

ok so 130*39=5070, which is close to 5200 which is the number of regularization images. Is that numb

ok so 130*39=5070, which is close to 5200 which is the number of regularization images. Is that number similarity a coincidence or is that what I'm aiming for, for the number of repeats multiplied by the number of images to be about equal to the number of reg images?

ZZax ok so 130*39=5070, which is close to 5200 which is the number of regularization ...

Furkan Gözükara SECourses•12/26/23, 10:58 PM

ye i calculated that way

Furkan Gözükara SECourses•12/26/23, 10:59 PM

but it could lead to overtraining

Furkan Gözükara SECourses•12/26/23, 10:59 PM

since it will be equal to

Furkan Gözükara SECourses•12/26/23, 10:59 PM

130 * 39 * 2 steps

Furkan Gözükara SECourses•12/26/23, 10:59 PM

usually 15 images with 150 repeat yields good results even little bit overtrained for realism

FFurkan Gözükara SECourses good working

AIGambino•12/26/23, 11:02 PM

AAIGambino Click to see attachment

Furkan Gözükara SECourses•12/26/23, 11:22 PM

must be out of vram

Furkan Gözükara SECourses•12/26/23, 11:22 PM

show task manager

Furkan Gözükara SECourses•12/26/23, 11:22 PM

gpu

Furkan Gözükara SECourses•12/26/23, 11:22 PM

you know i can connect your pc and fix your python installaating and everything with private lecture

Furkan Gözükara SECourses•12/26/23, 11:22 PM

or you can follow this tutorial and instrall everything properly https://youtu.be/-NjNy7afOQ0

YouTubeSECourses

Essential AI Tools and Libraries: A Guide to Python, Git, C++ Compi...

Welcome to the Pillars of AI Application Libraries: How to Install Tutorial!

Are you ready to embark on a journey of installing AI libraries and applications with ease?

In this video, we'll guide you through the process of installing Python, Git, Visual Studio C++ Compile tools, and FFmpeg on your Windows 10 machine.

We'll also show you ho...

Essential AI Tools and Libraries: A Guide to Python, Git, C++ Compi...

Zono50•12/27/23, 12:32 AM

What is the most effective method of inpainting now?

ZZono50 What is the most effective method of inpainting now?

Furkan Gözükara SECourses•12/27/23, 12:47 AM

can you eloborate more?

Zono50•12/27/23, 12:48 AM

I see people making many crazy cursed images on Facebook. I try on stable diffusion and just doesn't work, like this one of krampus and pit bulls

Zono50•12/27/23, 12:48 AM

I assume they are inpainting the details

Original message was deleted

Furkan Gözükara SECourses•12/27/23, 12:49 AM

yes mask the area in auto1111 and inpaint

Furkan Gözükara SECourses•12/27/23, 12:49 AM

i think they are doing that

Zono50•12/27/23, 12:50 AM

Do you recommend inpaint models, controlnet inpainting or sdxl?

ZZono50 Do you recommend inpaint models, controlnet inpainting or sdxl?

Furkan Gözükara SECourses•12/27/23, 12:51 AM

i think just masking and inpainting with good prompt should work for that level of image

Furkan Gözükara SECourses•12/27/23, 12:51 AM

but you can try sd 1.5 controlnet inpainting nodes too

FFurkan Gözükara SECourses must be out of vram

AIGambino•12/27/23, 3:11 AM

AIGambino•12/27/23, 3:13 AM

https://pastebin.pl/view/23f066af

ttttt - Pastebin

Pastebin.pl is a website where you can store code/text online for a set period of time and share to anybody on earth

paleolithic•12/27/23, 3:51 AM

Need some help, need to ideate on what the background should look like. Is there something in control net I can use to ideate some background ideas?

Cc: @Dr. Furkan Gözükara

Darth Teethius•12/27/23, 5:34 AM

Anyone recommend a good lora dim/alpha for realistic sdxl models? Is 128/1 still best or has everyone moved to the dreamboat extract mode now?

FFurkan Gözükara SECourses usually 15 images with 150 repeat yields good results even little bit overtraine...

Baran•12/27/23, 11:10 AM

for one epoch?

BBaran for one epoch?

ZaxOP•12/27/23, 11:39 AM

if I'm correctly understanding what he's saying, an "epoch" is a "full run" of training on your data set. So multiple epochs constitutes multiple runs of training on your data set. This is functionally the same thing as doing repeats for images. so your total number of steps is [training images] x [repeats] x [epochs], and take that and double it if you have regularization images. So doing 15 training images with 150 repeats and 1 epoch is 2250 steps (or 4500 with reg images). If you instead did 15 training images with 10 repeats for 15 epochs, that would also be 2250 steps (4500 with regs).

Max•12/27/23, 12:48 PM

guys I have a question. Why does it give me a different output every time I copy a prompt from civitai and paste it into sdxl with the same settings? Shouldn't he give the same? Do I have to install like texture inversion? Help me

Lorelai AI•12/27/23, 1:21 PM

@Dr. Furkan Gözükara I am playing around with different settings and one model is giving me issues in training a Laura. Most go smooth based on your video. I've tried more pictures, less pictures, more repeats, etc. I'm trying a higher learning rate right now. My go to settings are usually 2 batches(4090 doesn't appear to handle 3 or more on AdamW) 15 repeats and 15 Epochs. Default learning rate. This has worked awesome with 14 images or I've had great results with 8 repeats, 15 epochs, default learning rates, but 29 images. My subject is a weightlifter and I'm training with SDXL Base. Any tips?

Ppaleolithic Need some help, need to ideate on what the background should look like. Is there...

Furkan Gözükara SECourses•12/27/23, 1:56 PM

what are you trying right now?

AAIGambino Click to see attachment

Furkan Gözükara SECourses•12/27/23, 1:57 PM

something else is looking wrong

Furkan Gözükara SECourses•12/27/23, 1:57 PM

not using vram

DDarth Teethius Anyone recommend a good lora dim/alpha for realistic sdxl models? Is 128/1 still...

Furkan Gözükara SECourses•12/27/23, 1:59 PM

it need further testing

Furkan Gözükara SECourses•12/27/23, 1:59 PM

but i tested lora extract

Furkan Gözükara SECourses•12/27/23, 1:59 PM

128x128 and 192x192 and 256x256 almost same

Furkan Gözükara SECourses•12/27/23, 1:59 PM

when you extract

Furkan Gözükara SECourses•12/27/23, 1:59 PM

out of dreambooth model

BBaran for one epoch?

Furkan Gözükara SECourses•12/27/23, 1:59 PM

yes. because 150 repeating making it 150 epoch actually

Furkan Gözükara SECourses•12/27/23, 1:59 PM

since every training image is 150 times trained

MMax guys I have a question. Why does it give me a different output every time I copy...

Furkan Gözükara SECourses•12/27/23, 2:14 PM

seed matters

FFurkan Gözükara SECourses seed matters

Max•12/27/23, 2:15 PM

but i put them

LLorelai AI @Dr. Furkan Gözükara I am playing around with different settings and one model i...

Furkan Gözükara SECourses•12/27/23, 2:16 PM

i noticed same. sdxl realistic vision was undertrained at my settings

MMax but i put them

Furkan Gözükara SECourses•12/27/23, 2:16 PM

could have extra loras etc

FFurkan Gözükara SECourses i noticed same. sdxl realistic vision was undertrained at my settings

Lorelai AI•12/27/23, 2:50 PM

I used a Learning rate of .001, thew results were much better! .001 might be high but because I'm doing parody I don't really need exact I just need to capture their likeness and make them something else

Enlil Wolf•12/27/23, 3:23 PM

@Dr. Furkan Gözükara - Dr, when you get the chance, please check the DM

LLorelai AI I used a Learning rate of .001, thew results were much better! .001 might be hig...

Furkan Gözükara SECourses•12/27/23, 3:24 PM

ye it changes according to settings

B0nd•12/27/23, 4:35 PM

When training for SDXL DreamBooth or Lora in KoyhaSS what is generally the best way, for likeness and flexibility?

With or without regularization images?
With or without captioning the dataset?

FFurkan Gözükara SECourses you need to use our 16gb config

DigitalArt_Lover•12/27/23, 5:41 PM

I need to know, the 16gb config, can i get it from you by any chance?? ..i would buy the Patreon way long ago, but for my country i could pay, because of the payment method restrictions.

BB0nd When training for SDXL DreamBooth or Lora in KoyhaSS what is generally the best ...

Furkan Gözükara SECourses•12/27/23, 7:54 PM

it depends purpose. i made a good comparison here recently

Furkan Gözükara SECourses•12/27/23, 7:55 PM

https://civitai.com/articles/3510/dreambooth-training-on-sdxl-anime-model-and-comparing-ground-truth-regularization-images-effect

DreamBooth Training on SDXL Anime Model & Comparing Ground Truth Re...

It was requested of me to test ground truth Regularization / Classification images effect during Stable Diffusion XL (SDXL) DreamBooth training. So...

DDigitalArt_Lover I need to know, the 16gb config, can i get it from you by any chance?? ..i would...

Furkan Gözükara SECourses•12/27/23, 7:55 PM

hello. sure i accept crypto and other methods too

ok so 130*39=5070, which is close to 5200 which is the number of regularization images. Is that numb

Similar Threads