you cant use multiple gpu for a single training with automatic1111 or existing scripts. it is much h
you cant use multiple gpu for a single training with automatic1111 or existing scripts. it is much harder to do
Fyes it uses number of class images you set for per training image
JWhat is the program called?
GRunPod
Fi have a very good tutorial for runpod actually multiple
FYouTubeSECourses - Software Engineering Courses
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...
https://www.patreon.com/SECourses...FYouTubeSECourses - Software Engineering Courses
RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. How to install and start using #DreamBooth extension on #RunPod SD Web UI template. 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
DreamBooth extension: ht...
https://www.patreon.com/SECoursesDreamBooth extension: ht...
Gthis does not look good
Jok i feel so stupid right now.
I just saw in the last video that propts were separated with a comma.
This has improved my results so much.
I just saw in the last video that propts were separated with a comma.
This has improved my results so much.
Fit looks good low loss rate
Si keep getting this out of memory thing after I have trained two different model. Tried to restart and clear it, however it reappears very quickly, and i cant train any new model now. Is there a way to fix it?
Smoreover i realise pytorch 2 doesnt work with xformer with my 1080 Ti, is there a way to it?
Fcurrently none. you cant use yet
Ftry --lowvram
Sis there a way i could reset it so i can start training model again? not sure why it is jamming those vram all the time
Fclose cmd
Fand start again
GSo running via RunPod, I don't want to set up Dreambooth every time I want to use it and I don't want to pay per hour for having it idle, Do I just download the folder and upload it each time ?
Fif you dont delete your pod it remains
Gyes but it's 1c per hour. I forgot about my runpod account for 3 months and lost around 50 dollars XD
Fwell i think you have to reinstall
Fif you make a new pod
Gah okay
Gno worries then
Sthanks. It worked at the beginning but once i try to upscale a photo then it got out of memory again. So --lowvram is the only solution for it now?
JI have a problem that when I set the resolution of the images to be generated above 512p. Suddenly too many of the described things appear or have too many body parts.
Same setting only maximum resolution. (The higher the resolution, the worse the effect).
example:
promt; foto, one, dancig, woman
Same setting only maximum resolution. (The higher the resolution, the worse the effect).
example:
promt; foto, one, dancig, woman
J512p
J
GSuccess!!!!
Galthough facial features are a bit off and the eyes are broken
G
GI think better settings on runpod will be much better
J(2048)
GYour image is to big
JSo you have to upscaling for large images?
Gcorrect
JWe should create a basic tutorial for basics. (The smal stuf)+
Gyou can only use 512x512 on 1.5 and 720x720 on 2
Fwell unfortunately you cant upscale much :/
Fthis is expected
Fyou can try highres fix
Fyou need to add beautifying words
Ffor example photo of and photoshop beautifies significantly
Gquick question , on runpod with the models that are with the stable-diffusion is it still needed to upload the 1.5v pruned model?
G
Fi prefer that for training
Fthat one better
Fuse wget downloads very quickly
Gwhere can I get the link?
