Software Engineering Courses (SECourses)•3y ago

So I'm still watching this tutorial and in the minute 8:15 he selects a model from a folder call "0

So I'm still watching this tutorial and in the minute 8:15 he selects a model from a folder call "0 models". But I don't have it. Do you know how can I get those models? https://www.youtube.com/watch?v=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

https://bit.ly/SECoursesDiscord

If I have been of assis...

Fflowwolf terminal > make sure no venv is active > `cd \kohya_ss\ [whereever it is]` `pi...

▲

▲•6/2/23, 4:49 AM

thanks i'm gonna try this, cause i havent solved yet

▲

▲•6/2/23, 4:50 AM

however, beware of latest nvidia driver update. it slow down my SD to terrible speed

▲

▲•6/2/23, 4:50 AM

i had to roll back driver

RRclemente27 So I'm still watching this tutorial and in the minute 8:15 he selects a model fr...

▲

▲•6/2/23, 4:54 AM

there's plenty of models on hugginface or civitAI

▲▲however, beware of latest nvidia driver update. it slow down my SD to terrible s...

flowwolf•6/2/23, 5:11 AM

is fine here

Fflowwolf is fine here

▲

▲•6/2/23, 5:21 AM

runned your code
had to choose "yes" to tons of question on uninstall
reinstalled
still not working
i dunno what to do anymore

FFurkan Gözükara SECourses testing your template @ashleyk right now

ashley•6/2/23, 5:23 AM

How did it go?

▲

▲•6/2/23, 5:23 AM

not only, now my stable diffusion runs at 1/its

▲

▲•6/2/23, 5:24 AM

i guess i shut off my computer for today and go to ride my bycicle in nature

shameforest•6/2/23, 6:18 AM

Hey Guys!
Do you recommend any good tips, workflows or strategies on how to create specific graphic novel framing and images? Not on an amateur level, just creating some random comic images, but with specific camera framing and characters with specific facial expressions, etc?
If you have any ideas or material to recommend, that would be very helpful. Thanks!

Enlil Wolf•6/2/23, 11:16 AM

Hey there guys, anyone here facing any issues where you have already uploaded pre-set classification images but Dreambooth starts to create a batch of new images? It shouldn't be happening as the process itself is so slow.

EEnlil Wolf Hey there guys, anyone here facing any issues where you have already uploaded pr...

ashley•6/2/23, 11:28 AM

That happens when your directory to the classification images is incorrect

ashley•6/2/23, 11:29 AM

Check that you never made a tying error when entering it

felixsanz•6/2/23, 11:39 AM

@Dr. Furkan Gözükara haven't you find that training a face on realistic vision model is usually a very bad idea? i don't know if i'm doing it wrong, etc. but my results are horrible always

Aashley How did it go?

Furkan Gözükara SECourses•6/2/23, 11:46 AM

it asked password to connect jupyter i did a new installation :d

Ffelixsanz @Dr. Furkan Gözükara haven't you find that training a face on realistic vision m...

Furkan Gözükara SECourses•6/2/23, 11:47 AM

i am working on a new workflow

Furkan Gözükara SECourses•6/2/23, 11:47 AM

generated over 8k 768x1024 class images. trying to remove nsfw automatically haha

felixsanz•6/2/23, 11:47 AM

on realistic vision?

FFurkan Gözükara SECourses generated over 8k 768x1024 class images. trying to remove nsfw automatically hah...

felixsanz•6/2/23, 11:47 AM

use the nsfw filter that roop repo has added recently

Ffelixsanz use the nsfw filter that roop repo has added recently

Furkan Gözükara SECourses•6/2/23, 11:51 AM

i found a nice .net library

Furkan Gözükara SECourses•6/2/23, 11:51 AM

making chatgpt work haha

NNom anyone here know how to make class images work with the dev branch of dreambooth...

Furkan Gözükara SECourses•6/2/23, 11:52 AM

i installed it on runpod yesterday.

Furkan Gözükara SECourses•6/2/23, 11:52 AM

skip install on auto1111 begining and also use latest auto1111

felixsanz•6/2/23, 11:52 AM

Ccode With my 4090 and torch 2 it takes around ~50 seconds to generate two x 2048x102...

Furkan Gözükara SECourses•6/2/23, 11:52 AM

wow huge diff

Furkan Gözükara SECourses•6/2/23, 11:54 AM

i love c# :d

Furkan Gözükara SECourses•6/2/23, 11:55 AM

failed to move this

Aashley That happens when your directory to the classification images is incorrect

Enlil Wolf•6/2/23, 11:55 AM

Well, I just used 'copy path' and paste it after giving the /

So, that shouldn't be happening :sad:

FFurkan Gözükara SECourses failed to move this 😄

felixsanz•6/2/23, 11:55 AM

that's not nsfw

Furkan Gözükara SECourses•6/2/23, 11:55 AM

i will share dataset on patreon

Ffelixsanz that's not nsfw

Furkan Gözükara SECourses•6/2/23, 11:55 AM

i know :d

Furkan Gözükara SECourses•6/2/23, 11:55 AM

but abomination

Furkan Gözükara SECourses•6/2/23, 11:55 AM

felixsanz•6/2/23, 11:56 AM

that's why i think regularization images is a bad idea

Furkan Gözükara SECourses•6/2/23, 11:58 AM

almost half is nswf lol

felixsanz•6/2/23, 11:59 AM

why you didn't include "nsfw" in the negative prompt? lmao

Furkan Gözükara SECourses•6/2/23, 12:00 PM

need to do some clearing

Ffelixsanz why you didn't include "nsfw" in the negative prompt? lmao

Furkan Gözükara SECourses•6/2/23, 12:00 PM

could have work maybe :d

Furkan Gözükara SECourses•6/2/23, 12:00 PM

this is 768x1024 so have some repeating problem

felixsanz•6/2/23, 12:01 PM

you could use sd 2.1 to generate them

Furkan Gözükara SECourses•6/2/23, 12:02 PM

Ffelixsanz you could use sd 2.1 to generate them

Furkan Gözükara SECourses•6/2/23, 12:02 PM

sd 2.1

Furkan Gözükara SECourses•6/2/23, 12:09 PM

very interestingly this model is putting a noise color to images

Furkan Gözükara SECourses•6/2/23, 12:09 PM

Furkan Gözükara SECourses•6/2/23, 12:09 PM

this exists in majority

Furkan Gözükara SECourses•6/2/23, 12:10 PM

Furkan Gözükara SECourses•6/2/23, 12:12 PM

but the variety of the model excellent

Furkan Gözükara SECourses•6/2/23, 12:16 PM

interesting fp32 version dont have that

Marabunta•6/2/23, 12:36 PM

Can you create an advanced dreambooth video for classification.
I know the video with ohwx man.
I mean a new one where it learns "jumping man mid air towards a hanging rope on the moon" kind of thing.
So how it not only learns a new dataset like a face and adapt the rest it, but also a whole new situation where the rest of dataset can adapt to - while learning multiple new keywords at the same time(is that even possible?).
And the last thing is classification images - if I have a "jumping man towards a hanging rope" - image(like indiana jones) can I create a textfile with the same name as the image, with the keywords for that image? if yes, how do i tell sd to use that textfile - and many other files too(maybe even classification by folder structure? /jumping/man/rope/indianajones001.jpg).
Can I only train the same keywords, so like only "jumping man towards a hanging rope" or can I also mix it with images of only "jumping man", "rope", "jumping man on on the moon"...
I hope the explanation is clear enough. Because I was struggling to find pictures for a situation and in detail classify it (do I use ohwx jumping man, or use the full description for that photo).

To know and understand this would boost my ability to use sd 100x better

So I'm still watching this tutorial and in the minute 8:15 he selects a model from a folder call "0

Similar Threads