I'm still stumped on this one
I'm still stumped on this one
Fno torch means torch not installed
Fbut kohya gui installs it
Fdid you do a fresh install?
Fi told you to do 
SAny update to training SDXL on kaggle since they added more CPUs? or should it still work. Just attempted and got error message but also was trying to use juggernautxl from huggingface as training model
LNo I was a bad boy and wanted to be lazy. I'll do it
Lhaha
Fit is mandatory since we reinstalled pythoın
Fit is working
Fbut base version
Fjuggernautxl somehow requires more vram
Sah disappointing, but base version training works well enough on juggernaut so that should be fine
Fperhaps you can download it into workspace as safetensors
Fand give its path
Shttps://i.imgur.com/56CjzyV.png any ideas how to fix this? As far as I'm aware I'm not running lowvram or medvram options
Imgur
SThats the only results I'm seeing when searching this issue
Fremove --medvram
F--lowvram
Falso for controlnet
Fthey have development branch
Fgit checkout dev
Fon automatic1111
SAre the options for that somewhere else? thats my webui-user.bat
Fok you dont have them
Fso try to swtich dev branch
Swill try
SFixed it, thank you
Bso i used the auto_1111 from the link in all_files_v3 but i cant start webgui
Fshow me error
Fyou will start from webui-user.bat you know that right?
Finside stable-diffusion-webui
Bmy command prompt closed when i clicked snipping too
Fi will go to sleep
Fif you open anydesk immediately
Fi can check
Byea i got the screen shot
B
Fwhat is your GPU?
B
Fthis doesnt show gpu
Fon the internet it says NVIDIA® GeForce RTX™ 3050 (8 GB GDDR6 dedicated)
Fsadly i have to go
Fif you have AMD
Fyou need to add
F--skip-torch-cuda-test
Bdo i type that command in?
Fand remove
Fxformers