Software Engineering Courses (SECourses)•17mo ago

so you recommend I train 10,000 images instead of 1000 right? but it wont overtrain? like you are m

so you recommend I train 10,000 images instead of 1000 right? but it wont overtrain? like you are mentioning

GGomezBro so you recommend I train 10,000 images instead of 1000 right? but it wont overt...

Furkan Gözükara SECourses•9/1/24, 10:33 AM

yes but as i said

Furkan Gözükara SECourses•9/1/24, 10:33 AM

dataset matters

GomezBroOP•9/1/24, 10:33 AM

whats that?

GomezBroOP•9/1/24, 10:33 AM

the quality of the images?

NNeuromasters it writes errors, do I have to put in an experimental file?

Furkan Gözükara SECourses•9/1/24, 10:33 AM

restart computer reinstall

Furkan Gözükara SECourses•9/1/24, 10:33 AM

if error remains disable antivirus etc

GGomezBro the quality of the images?

Furkan Gözükara SECourses•9/1/24, 10:34 AM

please check style training video : https://youtu.be/m-UVVY_syP0

YouTubeSECourses

Training Midjourney Level Style And Yourself Into The SD 1.5 Model ...

RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. Get your own #Midjourney level model for free via Style Teaching. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

Playlist of #StableDiffusion Tutorials, Au...

Training Midjourney Level Style And Yourself Into The SD 1.5 Model ...

Furkan Gözükara SECourses•9/1/24, 10:34 AM

the logic is same

TThereiver Hey, does anyone have a prodigy config for flux? Back in the day we’d set the wa...

Furkan Gözükara SECourses•9/1/24, 10:34 AM

it uses more vram

what gpu u have?

GomezBroOP•9/1/24, 10:35 AM

So I should do the 10K images over the 1K then right? So final question is how many days would it take to train? And whats the fastest VPS you recommend? I want to do this ASAP

GomezBroOP•9/1/24, 10:35 AM

Ill pay for 4 or 8 GPU rental

Thereiver•9/1/24, 10:36 AM

I’m using a 4090

GGomezBro So I should do the 10K images over the 1K then right? So final question is how m...

Furkan Gözükara SECourses•9/1/24, 10:36 AM

Massed compute go with 4x gpu

Furkan Gözükara SECourses•9/1/24, 10:36 AM

1.25 usd hour

GomezBroOP•9/1/24, 10:36 AM

How do I calculate the hours it will take?

Furkan Gözükara SECourses•9/1/24, 10:36 AM

But I suggest you to get a private lecture before such big traininh

GomezBroOP•9/1/24, 10:37 AM

whats that

TThereiver I’m using a 4090

Furkan Gözükara SECourses•9/1/24, 10:37 AM

I don't know if work on that

Thereiver•9/1/24, 10:37 AM

It does work, just need the right batch size

GGomezBro whats that

Furkan Gözükara SECourses•9/1/24, 10:37 AM

I give 1 to 1 lecture

TThereiver It does work, just need the right batch size

Furkan Gözükara SECourses•9/1/24, 10:37 AM

Batch size 1 best

Thereiver•9/1/24, 10:38 AM

For quality I agree

Furkan Gözükara SECourses•9/1/24, 10:38 AM

Batch size 2 made like %25 improvement

Furkan Gözükara SECourses•9/1/24, 10:38 AM

Rest didn't do

Furkan Gözükara SECourses•9/1/24, 10:38 AM

I tasted on a6000

GomezBroOP•9/1/24, 10:40 AM

even batch size 4 works on A6000 bros I tested it today

Thereiver•9/1/24, 10:40 AM

Thanks - will keep that in mind! Yeah it would be awesome if you could make a config for prodigy, I was able to train the darkest dungeon sdxl Lora in probably 1/10th of the time

GomezBroOP•9/1/24, 10:40 AM

RANK 5

TThereiver Thanks - will keep that in mind! Yeah it would be awesome if you could make a co...

Thereiver•9/1/24, 10:40 AM

I will have a look to see if I can find my old config

GomezBroOP•9/1/24, 10:40 AM

I got it to over 30+ iterations

GomezBroOP•9/1/24, 10:41 AM

it has a lot of VRAM i figured what the hell and tried it

Thereiver•9/1/24, 10:43 AM

Nice one man

Thereiver•9/1/24, 10:51 AM

Found my old prodigy config - will do some testing with flux dev and let you know if it's any good. Note I'm training a style, so the dims will be higher

TThereiver Found my old prodigy config - will do some testing with flux dev and let you kno...

GomezBroOP•9/1/24, 10:58 AM

brother i want to do that too, how many images will u train?

GomezBroOP•9/1/24, 10:58 AM

please do share infos

Thereiver•9/1/24, 10:58 AM

350 images in dataset roughly

GomezBroOP•9/1/24, 10:58 AM

i want to train an NFT collection into a style

Thereiver•9/1/24, 10:59 AM

This was my lora https://civitai.com/models/65324/darkestdungeon

DarkestDungeon - v2.1 | Stable Diffusion LoRA | Civitai

v2.1 Updated 19/5/23 If you enjoy the models I create, please consider buying me a coffee! Ko-fi I am pleased to announce the release of DarkestDun...

GomezBroOP•9/1/24, 10:59 AM

why so little if you dont mind me asking bro?

Thereiver•9/1/24, 10:59 AM

350 is quite a lot for a lora tbf

TThereiver This was my lora https://civitai.com/models/65324/darkestdungeon

GomezBroOP•9/1/24, 10:59 AM

thats a while back cool

GomezBroOP•9/1/24, 10:59 AM

Dr just told me to use more

GomezBroOP•9/1/24, 10:59 AM

to do 10K over 1K images

Thereiver•9/1/24, 11:00 AM

I would say 10k for a model not a lora but thats my opinion. 2000 for a style lora is a good amount. but you can get away with less

Thereiver•9/1/24, 11:01 AM

also mine: https://civitai.com/models/124498/darkest-dungeon-sdxl-lora

Darkest Dungeon SDXL Lora - v1 - Main Release | Stable Diffusion Lo...

I am pleased to announce the release of DarkestDungeon SDXL Lora What is it? Darkest Dungeon SDXL is a lora intended to capture the artstyle from '...

Thereiver•9/1/24, 11:01 AM

It's only right I make one for flux

TThereiver I would say 10k for a model not a lora but thats my opinion. 2000 for a style lo...

GomezBroOP•9/1/24, 11:07 AM

the thing is i want to make the quality of that NFT very closely, so close that it looks like it with the prompt i create

GomezBroOP•9/1/24, 11:07 AM

so i want it to be very accurate and high quality

Thereiver•9/1/24, 11:08 AM

Yeah it's doable. You just need a consistent dataset

Thereiver•9/1/24, 11:09 AM

Mine for instance has about 8 different concepts within

so you recommend I train 10,000 images instead of 1000 right? but it wont overtrain? like you are m

Similar Threads

Similar Threads

Similar Threads