Software Engineering Courses (SECourses)•3y ago

@Furkan Gözükara SECourses After a week of struggling to understand and to get things to run and not

@Furkan Gözükara SECourses After a week of struggling to understand and to get things to run and not crash in the middle of training I have finally managed to get some acceptable results. Thanks to your tutorials and advice. Now on to style training...

Zono50•4/28/23, 3:18 PM

if i write multiple words in my prompt, should i surround them with parenthesis?

Zono50•4/28/23, 3:18 PM

like i'm doing an indian, and i wrote "face with intricate warpaint"

Joyfie•4/28/23, 3:29 PM

Interesting, just watched the LoRA training guide, there's a lot of talk about Network Rank, it seems that lower network ranks seem to work well for some people, whereas the higher ranks seem preferable in other instances

Joyfie•4/28/23, 3:30 PM

I experimented a bit on my face, and I feel the higher network rate is much better, especially for finer details and 'ideas'

weilu•4/28/23, 3:38 PM

I just watched the tutorial about Lora, if I want to train an illustration style, how should I set the Dimension and Alpha?

Isaac Brock•4/28/23, 3:54 PM

where is the tutorial for making a lora?

Isaac Brock•4/28/23, 3:54 PM

do i have to use dreambooth?

Isaac Brock•4/28/23, 3:54 PM

i want to be able to use any model, so I dont want to make a dreambooth model

DrSteve N3TBJ•4/28/23, 3:57 PM

@Furkan Gözükara SECourses Doctor G, I want to follow along with your latest video with the Kohya LoRA Stable Diffusion training. I'm on GNU/Linux, will that make a difference? I don't think I'd use Visual Studio for example. Thank you for all of the work you put into every one of your videos!!!

Eggo•4/28/23, 4:03 PM

Hey! Watching your new Lora video now. Although, I thought using dream booth extension for Lora training was better? Should I continue using dream booth, or Kohya gui?

FFurkan Gözükara SECourses i just tested and it works

pat._.fenis•4/28/23, 4:13 PM

What version of the LDSR model are you using ? Is xformers required in the command line arggs for LDSR?

Zono50•4/28/23, 4:39 PM

I totally forgot i had a new installation and forgot to redownload VAE

Zono50•4/28/23, 4:39 PM

no wonder my stuff ain't been looking top notch

TheRustingDeadAdmin•4/28/23, 4:45 PM

the only decent images i get are from img2img, never gotten any decent results from training. GTX 1070 8GB

DrSteve N3TBJ•4/28/23, 4:50 PM

@Furkan Gözükara SECourses Odd, I'm following along with the Kohya LoRA video and I don't seem to have the same interface? I don't see the Network Rank and Network Alpha and I don't see anything about LORA on the interface? I don't know what I could have done wrong.

Isaac Brock•4/28/23, 4:53 PM

\stable-diffusion-webui\venv\lib\site-packages\torchvision\transforms\functional_tensor.py:5: UserWarning: The torchvision.transforms.functional_tensor module is deprecated in 0.15 and will be removed in 0.17. Please don't rely on it. You probably just need to use APIs in torchvision.transforms.functional or in torchvision.transforms.v2.functional.
?

DDrSteve N3TBJ @Furkan Gözükara SECourses Odd, I'm following along with the Kohya LoRA video an...

DrSteve N3TBJ•4/28/23, 4:58 PM

Never mind me. I am so dumb I was in the "DREAM BOOTH" tab NOT the Dreambooth LORA" tab, I guess I missed that part.

Zono50•4/28/23, 5:03 PM

Great Kohya video! Didn't realize that you didn't need captions unless you were training multiple styles

goody•4/28/23, 5:13 PM

is it possible to train Kandinsky w/ DB? not super clear to me on how much is model, how much is code

Isaac Brock•4/28/23, 5:25 PM

in the new lora tutorial you say to install the old torch i just uninstalled for 2

Isaac Brock•4/28/23, 5:33 PM

so i used torch 2 in the install for kohys

Lacerda613•4/28/23, 5:54 PM

@Furkan Gözükara SECourses thank you for latest video about Lora. I am trying to generate reg images, but they all look like sketches, not realistic portraits. I am using same model and VAE in video. Any advice?

Zono50•4/28/23, 6:02 PM

@Lacerda613 what model are you using?

Lacerda613•4/28/23, 6:05 PM

Realistic vision 2.0, like in the tutorial

DrSteve N3TBJ•4/28/23, 6:35 PM

Keep failing. I think my installations are borked, not right for my system/set-up.

LLacerda613 @Furkan Gözükara SECourses thank you for latest video about Lora. I am trying to...

TheRustingDeadAdmin•4/28/23, 6:47 PM

same thing happens to me

Joyfie•4/28/23, 7:53 PM

Any opinions on captioning for face TI training? Do you structure your captions like "Photograph of a man, wearing a red shirt, white background" - Or do you tend to just have a single string without commas?

Joyfie•4/28/23, 7:54 PM

It occurred to me a while ago that commas carry their own weights, and do have somewhat of an influence on generation

Joyfie•4/28/23, 7:54 PM

I guess the logic also applies to training

Isaac Brock•4/28/23, 8:05 PM

doesnt seem to matter for me

Isaac Brock•4/28/23, 8:05 PM

the BLIP uses one long sentence seems to be fine

Isaac Brock•4/28/23, 8:06 PM

i just confirm they are correct

Joyfie•4/28/23, 8:06 PM

Seems like it, I might just go through and delete all of my commas and run the training again, see if I get wildly different results

Joyfie•4/28/23, 8:06 PM

There's some people out there who place a lot of emphasis on captioning, a lot of people say it doesn't matter

Isaac Brock•4/28/23, 8:07 PM

and its correct 95%+ pretty cool

Isaac Brock•4/28/23, 8:08 PM

i bet in a couple years youll just be able to give it a few pictures of people and just tell it to make the best movie ever with them

TheRustingDeadAdmin•4/28/23, 8:11 PM

yeah, not very useful until then. I have a feeling its gonna have that same vibe that vhs tapes give when you watch them. Instantly recognizable that its stable diffusion.

Eggo•4/28/23, 8:12 PM

I have a quick question. For Lora training. I’m wondering why ohwx has to be used as instance token? If we’re training a person, can’t we just use their first and last name as the instance token, followed by the class token? Or does that not work as well ? (For instance prompt and class prompt I use file words)

Zono50•4/28/23, 8:18 PM

anyone seen the ai made pizza commercial yet?

Major Epidemic•4/28/23, 8:41 PM

Thanks for the kohya video @Furkan Gözükara SECourses you absolute legend.

Furkan Gözükara SECourses•4/28/23, 9:00 PM

Sorry guys I have been away will read messages

AAlexP @Furkan Gözükara SECourses After a week of struggling to understand and to get t...

Furkan Gözükara SECourses•4/28/23, 9:01 PM

Ty so much for the feedback

ZZono50 if i write multiple words in my prompt, should i surround them with parenthesis?

Furkan Gözükara SECourses•4/28/23, 9:04 PM

You don't need. () will make prompt strength 1.1

JJoyfie Interesting, just watched the LoRA training guide, there's a lot of talk about N...

Furkan Gözükara SECourses•4/28/23, 9:05 PM

Ye I also found 128 performed best with alpha value 1. I plan to test even higher values

Wweilu I just watched the tutorial about Lora, if I want to train an illustration style...

Furkan Gözükara SECourses•4/28/23, 9:07 PM

I would try with same settings and even with higher network rank and compare

IIsaac Brock where is the tutorial for making a lora?

Furkan Gözükara SECourses•4/28/23, 9:09 PM

Here the tutorial : https://youtu.be/TpuDOsuKIBo

YouTubeSECourses

Generate Studio Quality Realistic Photos By Kohya LoRA Stable Diffu...

#Kohya SS web GUI DreamBooth #LoRA training full tutorial. You don't need technical knowledge to follow this tutorial. In this tutorial I have explained how to generate professional photo studio quality portrait / self images for free with Stable Diffusion training.

Our Discord server