Software Engineering Courses (SECourses)•3y ago

not able to install / upgrade xformers now

Bblackorchid does anyone knows how to fix these errors?

Tashiketh•12/3/23, 3:39 PM

Which version of CUDNN did you install? It wont work with CUDA 12 you have to use CUDA 11.8 versions

WWalter white.not able to install / upgrade xformers now

Walter white.OP•12/3/23, 3:39 PM

i think its a read issue. ugh what to do

TTashiketh Which version of CUDNN did you install? It wont work with CUDA 12 you have to us...

blackorchid•12/3/23, 3:49 PM

Thank you, I'm using Torch 2.0.1+cu118

Zono50•12/3/23, 3:53 PM

@Dr. Furkan Gözükara INstalling manually on my linux machine, where do we put the models at? I don't see a models folder for pixart?

Zono50•12/3/23, 6:03 PM

how many gigs of vram does pixart use? everytime i run app.py, it says out of vram

mikemenders•12/3/23, 9:16 PM

I just trained with a very interesting set of images. Even though I selected the best photos, I could only find the best d_coef value for a quarter of them. I tried 0.85, it didn't come out strong. With 1.02 it was too strong, and even with 0.95. Finally, with 0.9 I managed to get a near perfect Lora. On another set of images, I managed to get one that turned out perfect the first time, even though there was nothing special about the photos. So there are still some exceptions where, even if the graph is good, you still need to polish the final result a bit.

Ssexyboi @Dr. Furkan Gözükara I saw on one of your linkedin posts that you mentioned you ...

Furkan Gözükara SECourses•12/4/23, 12:26 AM

yes i have : https://www.patreon.com/posts/very-best-for-of-89213064

TThibault We really need @Dr. Furkan Gözükara to do a tuto on that

Furkan Gözükara SECourses•12/4/23, 12:26 AM

i am following their repo

Furkan Gözükara SECourses•12/4/23, 12:26 AM

they havent released yet

WWalter white.@Dr. Furkan Gözükara i am currently going through your udemy course, and i got ...

Furkan Gözükara SECourses•12/4/23, 12:27 AM

that is very best PC

Furkan Gözükara SECourses•12/4/23, 12:27 AM

surely enough

Bblackorchid does anyone knows how to fix these errors?

Furkan Gözükara SECourses•12/4/23, 12:27 AM

just ignore them. tensorRT spits them out but works

ZZono50 @Dr. Furkan Gözükara INstalling manually on my linux machine, where do we put th...

Furkan Gözükara SECourses•12/4/23, 12:28 AM

you set default cache folder

Furkan Gözükara SECourses•12/4/23, 12:28 AM

and they get downloaded into there

Furkan Gözükara SECourses•12/4/23, 12:28 AM

hugging face cache folder

Furkan Gözükara SECourses•12/4/23, 12:28 AM

look at the runpod instructions it shows

ZZono50 how many gigs of vram does pixart use? everytime i run app.py, it says out of vr...

Furkan Gözükara SECourses•12/4/23, 12:28 AM

how much vram you have?

Zono50•12/4/23, 12:33 AM

ZZono50 12

Furkan Gözükara SECourses•12/4/23, 12:41 AM

can you try low vram file

Furkan Gözükara SECourses•12/4/23, 12:41 AM

i added

FFurkan Gözükara SECourses yes i have : https://www.patreon.com/posts/very-best-for-of-89213064

sexyboi•12/4/23, 12:41 AM

Perfect, thank you! Love your work

Zono50•12/4/23, 12:41 AM

I will thank you

Furkan Gözükara SECourses•12/4/23, 12:41 AM

if you are on linux just look at the hyper parameter when starting the app.py

Furkan Gözükara SECourses•12/4/23, 12:42 AM

let me know result please

Zono50•12/4/23, 1:05 AM

Will do

Walter white.OP•12/4/23, 3:16 AM

@Dr. Furkan Gözükara i am on the 2nd episode of your course on udemy, have you in latter courses also specified how to prompt stable diffusion to generate amazing images like how people do (chubby:0.4) something like that

DR.Siva•12/4/23, 4:02 AM

I am training my photos for sdxl lora today. I used ohwx woman for my wife , can i use same ohwx man ? or ohwx should be different ?

DR.Siva•12/4/23, 4:32 AM

ok i am using bbuk man for me and ohwx woman for my wife

DR.Siva•12/4/23, 4:33 AM

this should help me to create photos of us nboth i think

DR.Siva•12/4/23, 5:42 AM

Need clarity on sdxl Lora training kaggle notebook code

DR.Siva•12/4/23, 5:43 AM

There is different run code for Lora training after koyass webui run

DR.Siva•12/4/23, 5:43 AM

It's giving me old error

RelentlessDung•12/4/23, 6:33 AM

Hi, @Dr. Furkan Gözükara some questions about Kohya training for SD. I would like to train clothing based on a culture. Basically, using any checkpoint, I can just invoke them to wear the clothing with just a keyword. Would a Lora be the best option? Also, when training, how do you best caption the dataset so that the Lora does not remember the background, the face, color of the clothing, physical weight of the person, etc. I only want the clothing, and want to be able to determine the colors. How many images would be minimum for clothing? And what's the best way to caption? I have tried so many different ways, and haven't been very successful.

Benjamin•12/4/23, 8:43 AM

https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.7.0-RC

Automatic1111 is soon releasing 1.7.0

GitHub

Release 1.7.0-RC · AUTOMATIC1111/stable-diffusion-webui

How to use: wiki.
Features:

settings tab rework: add search field, add categories, split UI settings page into many
add altdiffusion-m18 support (#13364)
support inference with LyCORIS GLora netwo...

DR.Siva•12/4/23, 9:35 AM

ok i think i found the issue (possibly)

In json , there is add parameter ""additional_parameters": "--max_grad_norm=0.0 --no_half_vae --train_text_encoder","

removing --train_text_encoder solved teh sdxl_train.py error

i might be wrong , or this text shouldnt be removed.. @Dr. Furkan Gözükara should only confirm

DR.Siva•12/4/23, 10:19 AM

since we are doing lora training and this parameter seems to be for dreambooth , this is causing the error likely

DR.Siva•12/4/23, 10:21 AM

Kindly provide two different notebook files , one for dreambooth and another for lora , so that it wont confused , also the dreambooth_best json file keep overwriting the locally saved , whenever we need to rerun the kohyaass gui. Wish there was other option

DR.Siva•12/4/23, 10:26 AM

so when i use two different face trained lora , how am i supposed to use adetailer ? and that too correctly on male and female faces ?

WWalter white.@Dr. Furkan Gözükara i am on the 2nd episode of your course on udemy, have you i...

Furkan Gözükara SECourses•12/4/23, 10:32 AM

i think i had shown it but cant remember in which lecture it was

DDR.Siva I am training my photos for sdxl lora today. I used ohwx woman for my wife , can...

Furkan Gözükara SECourses•12/4/23, 10:32 AM

you should have seperate model

DR.Siva•12/4/23, 10:33 AM

like ?

RRelentlessDung Hi, @Dr. Furkan Gözükara some questions about Kohya training for SD. I would lik...

Furkan Gözükara SECourses•12/4/23, 10:33 AM

if you can collect 1000s of images

Furkan Gözükara SECourses•12/4/23, 10:33 AM

you can caption with llava

Furkan Gözükara SECourses•12/4/23, 10:33 AM

and really obtain improved model

Furkan Gözükara SECourses•12/4/23, 10:33 AM

i would do dreambooth and then extract lora instead of lora training

BBenjamin https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.7.0-RC ...

Furkan Gözükara SECourses•12/4/23, 10:33 AM

i am reading changes

DDR.Siva ok i think i found the issue (possibly) In json , there is add parameter ""addi...

Furkan Gözükara SECourses•12/4/23, 10:33 AM

use latest configs

Furkan Gözükara SECourses•12/4/23, 10:33 AM

they are made for latest gui

DR.Siva•12/4/23, 10:35 AM

my wife lora is ready.. "
I am creating mine now.
So lets say i use something like

<lora:wife:1>,(ohwx woman) standing with <lora:husband:1>,(bbuk man) in an island

DR.Siva•12/4/23, 10:35 AM

it will create two faces , in adetailer what prompt should we give ?

not able to install / upgrade xformers now

Similar Threads

Similar Threads

Similar Threads