Software Engineering Courses (SECourses)•2y ago

one trainer won't train what am I missing

AAIGambino one trainer won't train what am I missing

DarkView•4/4/24, 12:22 AM

tried a fresh install?

Rrafraf @Dr. Furkan Gözükara in kohya, can "Epoch" be set to "0" or empty if I set Max S...

Furkan Gözükara SECourses•4/4/24, 12:30 AM

never tried

Vvirginia @Dr. Furkan Gözükara can webui still be used?

Furkan Gözükara SECourses•4/4/24, 12:30 AM

yes

Kkareem9000 Can you explain what this means? do you mean do an original supir and then an up...

Furkan Gözükara SECourses•4/4/24, 12:30 AM

i mean do 1 training with original

Furkan Gözükara SECourses•4/4/24, 12:30 AM

1 with upscaled

Furkan Gözükara SECourses•4/4/24, 12:30 AM

and compare

AAIGambino one trainer won't train what am I missing

Furkan Gözükara SECourses•4/4/24, 12:31 AM

i dont know you can ask developer on their discord

Furkan Gözükara SECourses•4/4/24, 12:31 AM

he is super active

Furkan Gözükara SECourses•4/4/24, 1:09 AM

purchased Poco X6 pro and transferring all installed apps taking huge time

Pelolisu•4/4/24, 1:32 AM

poco x6 is good? i have a poco gt x3

Pelolisu•4/4/24, 1:35 AM

i think they only change screen type in x6 haha

AIGambinoOP•4/4/24, 1:45 AM

wheres the tier2 for onetrainer @Dr. Furkan Gözükara ?

AIGambinoOP•4/4/24, 1:45 AM

for 24gb 3090 cards

FFurkan Gözükara SECourses purchased Poco X6 pro and transferring all installed apps taking huge time

VAMP•4/4/24, 1:54 AM

You also like performance Doc...i have redmi k50 with dimensity 8100...works good..

VAMP•4/4/24, 1:54 AM

do you also root your devices doc?

VAMP•4/4/24, 1:54 AM

mine is rooted

PPelolisu poco x6 is good? i have a poco gt x3

Furkan Gözükara SECourses•4/4/24, 2:00 AM

i have poco x3 pro

Furkan Gözükara SECourses•4/4/24, 2:00 AM

upgrading to poco x6 pro - going to give mine to my wife

Furkan Gözükara SECourses•4/4/24, 2:00 AM

it is super powerful phone

AAIGambino wheres the tier2 for onetrainer @Dr. Furkan Gözükara ?

Furkan Gözükara SECourses•4/4/24, 2:00 AM

no more tier 2 after recent updates

Furkan Gözükara SECourses•4/4/24, 2:00 AM

all are tier 1

VVAMP do you also root your devices doc?

Furkan Gözükara SECourses•4/4/24, 2:01 AM

no i dont root

Furkan Gözükara SECourses•4/4/24, 2:01 AM

in past i did in few phones but not needing it

FFurkan Gözükara SECourses upgrading to poco x6 pro - going to give mine to my wife

Pelolisu•4/4/24, 2:03 AM

i want do this too git mine to my wife haha

Pelolisu•4/4/24, 2:03 AM

they always have the leftovers haha

PPelolisu they always have the leftovers haha

Furkan Gözükara SECourses•4/4/24, 3:28 AM

well my wife prefers getting clothes so she gets them

FFurkan Gözükara SECourses well my wife prefers getting clothes so she gets them 😄

VAMP•4/4/24, 4:08 AM

women usually love two things - clothes and jewellery

Morpheus SX343•4/4/24, 5:22 AM

Hi all,
has anyone yet tried to be more accurate while training specific body parts, e.g. abs and pecs of a trained man, etc.
Or legs, waist and bottom? Obviously you cannot always have the face on the pictures if you dont always wanna take full body shots, especially for lower bodyparts.

I'm wondering what is the possible way to use closeup pictures to be more detailled.
Right now I'm not labelling / captioning my pictures.
But I dont know how the trainer is supposed to know that this bottom belongs to "ohwx man" or similarly.

Has anyone experience if this specific task can be improved by only labelling those specific images exactly with using the instance prompt? "bottom of ohwx man from behind". Or stuff like this?

FFurkan Gözükara SECourses yes but i prefer kohya better

ArtChicken•4/4/24, 9:36 AM

Sebastianblood•4/4/24, 10:00 AM

@Dr. Furkan Gözükara I did not notice you were training up to 150 epoch for just 15 images. That makes me wonder what would be the ideal epoch to train up to for 4000 images?

Sebastianblood•4/4/24, 10:01 AM

Are your sample images on early epochs looking fried?

yunusos•4/4/24, 11:46 AM

@Dr. Furkan Gözükara , is there any LoRA model we can play with that shows the full implementation in pytorch ? I would like to see the layers , connections and everything to change the architecture and see the impact for instance

rafraf•4/4/24, 12:47 PM

@Dr. Furkan Gözükara testing out FullFinetune on Kohya and having issues with metadata processing. Do i have to run a script to convert my txt caption files? Do you have a video for this?

rafraf•4/4/24, 12:53 PM

OSError: [Errno 22] Invalid argument: 'E:\studio Dropbox\studio\ai\libs\SD\1models\1styles\gr4ph1cW_v001admw8bt_50wd14_rep5_bs6\n_20240404-145300.json'

rafraf•4/4/24, 12:53 PM

this is the error

AIGambinoOP•4/4/24, 1:49 PM

@Dr. Furkan Gözükara I followed your onetrainer tutorial with your preset and whenever I train a model its just broken. My current model wont even load in automatic1111