could it be trying to use your second gpu? amd?
could it be trying to use your second gpu? amd?
Fwith all models?
For certain model
Tfollowed your videos to the tee, any model
Trealistic vision v20 and sd 1.5 are my main ive been trying
Fso do a fresh install
Fyour installation is problem
Twhich part?
Ti am getting no errors
Fdont know
Ti thought they installed to venvs
Ti followed your videos exactly, even installation
Ti can do it again
Fhere my sd 1.5 man prompt
F
Fyou dont get this @TheRustingDeadAdmin ?
TI get those types of images yes
Tdid you pick out the sketch like ones
Fok it is working then
Fyou dont need
Mwell when i was setting up kohya it asked which gpu and i said all, also in the beggining when the ocmmand starts to run it says this
Fbase model produces good enough. with custom models they are very biased
Fthat can be reason
TThank you, ill reinstall again
Fcan you try with torch 2?
Mdo i have to uninstall and reinstall
Fno
Factivate venv
Finstall torch 2
M
Mshould i do multi gpu here?
SHi @Furkan Gözükara SECourses I would just like to let you know, I was able to follow the tutorial:
https://www.youtube.com/watch?v=zA4LksIVas8&t=306s
It worked, I trained my model, then I tried to launch i new pod today and it seens something got update because it no longer works anymore. Not even the first git pull on the root, it says something about one of the files, so if I do a git stash it lets me do the pull, but following everything as before don't wor anymore
https://www.youtube.com/watch?v=zA4LksIVas8&t=306s
It worked, I trained my model, then I tried to launch i new pod today and it seens something got update because it no longer works anymore. Not even the first git pull on the root, it says something about one of the files, so if I do a git stash it lets me do the pull, but following everything as before don't wor anymore
YouTubeSECourses
RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. How to install latest version of Automatic1111 Web UI and #DreamBooth on #RunPod. 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 #StableDiffusi...
https://www.patreon.com/SECoursesPlaylist of #StableDiffusi...
Adoes anybody have the original model of illuminatidiffusion and would like to share it ?
SI realized now it works, but now it launchs using 127.0.01:7860 instead of 0.0.0.0:3000 so the connect via http of the pod does not work and so i have to use the --share arg to be apple to access it
Sis there a way to change the ip it launches, I tried using --port 3000 but it also don't work with the runpod provided url
Z
Zusing kohya following docs video
Fyou dont need to do git pull
Fprobably you didnt kill previous instance
Fno first option
Fface is so high qualty
SOh I see, will try that! Thanks again
ZThanks doc
Zthat's using xformers 2
S@Furkan Gözükara SECourses have u tried using more augments apart from horizontal flip to see if they improve training?
Fwhich ones for example?
Si was thinking this, idk if makes sense using it tho, but there are plenty more https://pytorch.org/vision/stable/generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing
Fi think trying to improve training dataset would have more dramatic effect
Slike randomly grayscale, if its 5% of the times will not affect too much
Syeah i was thinking more in terms of preventing overfitting and still improving if possible
