this sd 2.1 training is really painful to make it work :d
this sd 2.1 training is really painful to make it work :d
Yits easy to make anime but the real question is how to do cool cartoon ilustration to me lol
Fi am trying to do it on runpod
0Use pixar or Disney
0Or superhero cartoons
LSo, do you have any idea why I would get empty black rectangles while training my model? it seems I have all correct versions and settings, no?
Fi am also getting on sd 2.1 atm :d
Fu are using
F0.0.14 and sd 1.5?
Fxformers 0.0.14
Lwhere can I see the sd version? I am not sure
Fwhich model you used as base when composing training_?
LOh, the f222 model, I believei it is based on 1.5
Lshould I use the official sd 1.5 ?
Fyes
Fcustom models are cooked
LShould I use the pruned version?
Fuse bigger size
LThank you! I will try
HWow, what is different?
Fye requires more vram
Fmade it work on runpod though :d
H
Frunpod recording finished
Fcovered all these
FGreetings everyone. In this video I am going to Show how to use Automatic1111 Web UI for stable diffusion tasks on runpod io like you are using it on your computer. I will cover many topics such as how to upload and download files quickly, how to delete directories, how to install and run extensions, how to quickly download and use custom models, how to do dreambooth training on sd 1.5 and sd 2.1 versions, how to do fine tuning via dreambooth, how to do textual inversion training. I will also explain how their pricing system works, how you can use bidding, how you can transfer files from pod to pod or from computer to pod and vice versa, how you can install custom other scripts such as Kohya gui. I will also demonstrate how you can use the new famous controlnet on runpod io.
F4 hours raw recording time to post process :/
HThis is a masterclass honestly.
HCould consider some Patreon only content maybe later on
Shorter vids on YouTube
Shorter vids on YouTube
HCan’t wait for AI to solve chopping up the timeline editing
Factually i found a way for it
Fit deletes no speech parts :d
Fi am using it now
HThat’s cool actually. Then you can just pause between key thoughts.
HIt syncs the video too?
Fyep
Fextremely useful
Fgood night hopefully see you tomorrow
AHola guys, I'm curious about when I use the workflow same as the video on Youtube, I got OOM. But I have a 3090 24GB. And then I change the settings, I close all the options that will cost VRAM. But it still OOM. It's teally strange...
A(Training the Dreambooth, not Lora)
HOn the A1111 extension? Maybe its an issue realted to that?
HI saw mentions of troubles in the extension today. But I am not sure
AYeah, on the A1111 extension, it's strange for that OOM on the 3090, even I use the lowest config of the extension...
Ncmd says so, can you help me, I must upgrade gpu right?
Cjust curious, but is anyone else have xformers break training on any commit of dreambooth? im using a 4090
CThe ratio I have of good eyes is amazing with my model.
Prompt: African woman
Prompt: African woman
C
Cdoes it look like midjourney to you guys?
Cprompts?
CLet see if you get the same thing as I do
C1girl, earrings, forehead, jewelry, lips, looking_at_viewer, ponytail, portrait, realistic, red_lips, solo 1girl, earrings, forehead, jewelry, lips, looking_at_viewer, ponytail, portrait, realistic, red_lips, solo
