Software Engineering Courses (SECourses)•3y ago

so i have collected a bunch of images and categorised, eg: medieval art, renaissance art, modern art

so i have collected a bunch of images and categorised, eg: medieval art, renaissance art, modern art, post mordern, digital etc etc. I want to train all these categories as a style in one single model. what would be the process for this?

bushOP•3/10/23, 3:25 PM

as in train all these catesgories with a different instance prompt for each category

EEnlil Wolf @Furkan Gözükara SECourses - What is the reason for this error and how to fix it...

Furkan Gözükara SECourses•3/10/23, 3:26 PM

reduce batch size

Furkan Gözükara SECourses•3/10/23, 3:26 PM

out of memory on gpu

Bbush so i have collected a bunch of images and categorised, eg: medieval art, renaiss...

Furkan Gözükara SECourses•3/10/23, 3:27 PM

filewords

Furkan Gözükara SECourses•3/10/23, 3:27 PM

explained in this video

Furkan Gözükara SECourses•3/10/23, 3:27 PM

https://www.youtube.com/watch?v=KwxNcGhHuLY

YouTubeSECourses - Software Engineering Courses

DreamBooth Got Buffed - 22 January Update - Much Better Success Tra...

Our Discord : https://discord.gg/HbqgGaZVmr. Newest update of DreamBooth extension of Automatic1111 brought huge quality and success improvement. 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 Stable Diffusion Tutorials...

bushOP•3/10/23, 3:27 PM

oh great, I'll check the video first

FFurkan Gözükara SECourses reduce batch size

Enlil Wolf•3/10/23, 3:28 PM

Could you check the DM as I don't wanna flood the main chat, since no one using colab

bushOP•3/10/23, 3:32 PM

oh and I've meaning to ask this. During dreambooth epic video, you used a classification folder with 2400 images of man, but then in the concepts tab you put "50 class images per instance image" and that generates a set number of class images again, why is that?

Deleted User•3/10/23, 3:44 PM

https://github.com/d8ahazard/sd_dreambooth_extension/commit/c295e9c1914ade4b3b1a57ccd564ca25fe34f523

GitHub

Gradio and new schedulers · d8ahazard/sd_dreambooth_extension@c295e9c

Deleted User•3/10/23, 3:45 PM

new optimizers: adamw (replaces adam), dadaptation sgd, dadaptation adamw, dadaptation adagrad

Deleted User•3/10/23, 3:51 PM

and ddpm noise scheduler added

Bbush oh and I've meaning to ask this. During dreambooth epic video, you used a classi...

Furkan Gözükara SECourses•3/10/23, 3:55 PM

no it wont generate again unless you need more than 2400

DDeleted User and ddpm noise scheduler added

Furkan Gözükara SECourses•3/10/23, 3:55 PM

i didnt notice much difference of it

EEnlil Wolf Could you check the DM as I don't wanna flood the main chat, since no one using ...

Furkan Gözükara SECourses•3/10/23, 3:55 PM

sure

Furkan Gözükara SECourses•3/10/23, 4:00 PM

haha animated dwayne johnson

Furkan Gözükara SECourses•3/10/23, 4:12 PM

what you think

bushOP•3/10/23, 4:27 PM

oh nice, another content creator NerdyRodent, he's been using something similar for a while now, but not sure if that also has the custom head movementsor not. Whats your method called?

Furkan Gözükara SECourses•3/10/23, 4:27 PM

Spline Motion Model. i plan a tutorial for this

Furkan Gözükara SECourses•3/10/23, 4:27 PM

the problem is resolution is not great

Furkan Gözükara SECourses•3/10/23, 4:28 PM

maybe i can use ultrasharp upscaler to upscale

bushOP•3/10/23, 4:29 PM

seems like it, it'll be good for like a floating head for your videos

bushOP•3/10/23, 4:30 PM

Have you tried the TRAIN IMAGIC ONLY optiomn in dreambooth extension?

Bbush Have you tried the TRAIN IMAGIC ONLY optiomn in dreambooth extension?

Furkan Gözükara SECourses•3/10/23, 4:31 PM

didnt

FFurkan Gözükara SECourses no it wont generate again unless you need more than 2400

bushOP•3/10/23, 4:32 PM

but, I uploaded 1700 reg images for 13 instance images and the slider at 100 but it generated another 400 and something more class images. Is it because i set the number of sample to generate at ''2''?

Bbush but, I uploaded 1700 reg images for 13 instance images and the slider at 100 but...

Furkan Gözükara SECourses•3/10/23, 4:33 PM

maybe it wasnt able to read all your class images

Furkan Gözükara SECourses•3/10/23, 4:33 PM

because this doenst make sense

bushOP•3/10/23, 4:34 PM

oh yes! that could be it. right.

Zono50•3/10/23, 5:11 PM

in stable diffusion, is there a good extension or method to basically overlay one image on top of another image?

gatsby•3/10/23, 6:14 PM

Does anyone know if there’s a fast way to install a trained model into Jupiter? It’s currently estimated to take hours for me to load it in. Just wondering if there’s a faster way?

Ggatsby Does anyone know if there’s a fast way to install a trained model into Jupiter? ...

Furkan Gözükara SECourses•3/10/23, 6:22 PM

yes

Furkan Gözükara SECourses•3/10/23, 6:22 PM

runpodctl

Furkan Gözükara SECourses•3/10/23, 6:22 PM

check chapters

Furkan Gözükara SECourses•3/10/23, 6:22 PM

https://youtu.be/QN1vdGhjcRc

YouTubeSECourses - Software Engineering Courses

Ultimate RunPod Tutorial For Stable Diffusion - Automatic1111 - Dat...

Sign up RunPod: https://bit.ly/RunPodIO. Our Discord : https://discord.gg/HbqgGaZVmr. This is the Grand Master tutorial for running Stable Diffusion via Web UI on RunPod cloud services. 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...

FFurkan Gözükara SECourses check chapters

gatsby•3/10/23, 6:34 PM

Thank you. Am checking now and after you download CTL you install a model via a url. If I’m trying to upload a model from my desktop, do I simply upload it straight into models without using the terminal? And it will automatically upload through CTL?

Mr Q•3/10/23, 6:52 PM

@Furkan Gözükara SECourses , i have a question, i was training a model in dreambooth using v1-5-pruned.ckpt as source checkpoint. after 150 epochs i stopped it and wanted to test it out, so i clicked on generate ckpt file, it gives me a savetensors file. when i use that model in the text2img it is only able to create the thing i trained, it is not combined with v1-5-pruned.ckpt anymore, is that right?

Ma7adi805•3/10/23, 6:55 PM

assert not shared.cmd_opts.lowvram, 'Training models with lowvram not possible'
AssertionError: Training models with lowvram not possible

assert not shared.cmd_opts.lowvram, 'Training models with lowvram not possible'
AssertionError: Training models with lowvram not possible

Ma7adi805•3/10/23, 6:57 PM

how can I fix this

MMr Q @Furkan Gözükara SECourses , i have a question, i was training a model in dream...

Mr Q•3/10/23, 7:00 PM

or do i need to use checkpoint merger afterwards?

MMa7adi805 ```assert not shared.cmd_opts.lowvram, 'Training models with lowvram not possibl...

justalurker•3/10/23, 7:05 PM

Check out: https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob/master/modules/hypernetworks/ui.py see if that's the fix you are looking for.

Ggatsby Thank you. Am checking now and after you download CTL you install a model via a ...

Furkan Gözükara SECourses•3/10/23, 7:19 PM

yes you can use runpodctl on your pc as in your pod

MMr Q @Furkan Gözükara SECourses , i have a question, i was training a model in dream...

Furkan Gözükara SECourses•3/10/23, 7:20 PM

not anymore ckpt. saved as safetensors

MMa7adi805 ```assert not shared.cmd_opts.lowvram, 'Training models with lowvram not possibl...

Furkan Gözükara SECourses•3/10/23, 7:20 PM

yes you cant train with lowvram

Furkan Gözükara SECourses•3/10/23, 7:20 PM

your options are cloud

Furkan Gözükara SECourses•3/10/23, 7:20 PM

runpod or google colab for example

MMr Q or do i need to use checkpoint merger afterwards?

Furkan Gözükara SECourses•3/10/23, 7:20 PM

you dont need

Jjustalurker Check out: https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob/master/m...

Furkan Gözükara SECourses•3/10/23, 7:20 PM

textual inversion dont work with lowvram i tested :d

Furkan Gözükara SECourses•3/10/23, 7:21 PM

hyper networks i didnt though

Furkan Gözükara SECourses•3/10/23, 7:21 PM

but probably wont be as well

Furkan Gözükara SECourses•3/10/23, 7:21 PM

generating base avatar for animation tutorial

so i have collected a bunch of images and categorised, eg: medieval art, renaissance art, modern art

Similar Threads

Similar Threads

Similar Threads