yes, used your config. used rundiffusion server with i think enough ram
yes, used your config. used rundiffusion server with i think enough ram
Cstill better results than when i train a quick model one dreamlook.ai but not as good as yours 
Frundiffusion server supports kohya?=
Fyou should train on runpod
Cyes
Cyes will try in the future
Fi doubt that they are properly supporting
Fyou can also message me your training dataset i can comment
Cok, sure. thanks so much
SI am just using diffusers to implement those things.
SThis is the basic result.
Fdiffusers is
Fvery different than automatic1111 or kohya
Fvery primitive
SIs there a way to use WebUI extension directly in diffusers?
SBecause I think what I am trying to do is outpainting more than inpainting, and seems inpainting models can't work properly(because they make changes from side.)
Fweb ui has api
Fyou can use it
SSo running WebUI and install extensions and access them via APIs?
SUnderstood.
are there any outpainting models available in diffusers?
are there any outpainting models available in diffusers?
Fi dont know sadly
SCould you explain some details of inpainting parameters and initialization?
SI think adjust noise can affect the final result.
Fsadly i only used automatic1111 implementation
M3.10.11
NHello @Dr. Furkan Gözükara I have been following your tutorial where you setup C++, ffmpeg and CUDA.
Unfortunately, I am on mac and CUDA setup is not possible. Should I use online Google colab as an alternative or do you have better suggestion ?
Unfortunately, I am on mac and CUDA setup is not possible. Should I use online Google colab as an alternative or do you have better suggestion ?
ADo you have any tutorials on using different controlnets? Like CN tile and upscaling?
EYes.
Actually after that I'm not able to start ngrok link.
Actually after that I'm not able to start ngrok link.
E
Vcan anyone help ;_;
LTraining 1.5 is a lot faster. first time trying it. 10 minutes vs 1 hour same data set except the checkpoint
IWhich is best way to get gif/mp4 output using stable diffusion with your own style trained model?
Fi suggest to use runpod
Fit is really good. i also have tutorials and auto installers
Fi have a big tutorial where i covered a lot of stuff you can look chapters
FYouTubeSECourses
I show how to install Automatic1111 Web UI & ControlNet extension installation from scratch in this video. Moreover I show how to make amazing QR codes and inpainting and out painting of ControlNet which are very similar to Photoshop generative fill and Midjourney zoom out. Furthermore, I explain and show what are Canny, Depth, Normal, OpenPose...
Fthis must be order error
Ffirst start ngrok
Fdo not click visit site
Fthen start web ui
Fthen click visit site
Fhello can you explain more
Fif you train 512x512 yes will be
Fsadly i dont know
LI hated how it came out but I believe it's because of the checkpoint I used
LMy only desire to try it is because the Lora availability on civit. But I'm going to stick to SDXL the trainings are an hour and I have good settings
Rmorning @Dr. Furkan Gözükara got a little issue on my laptop comfyui setup. any chance you could help me debug?
Fyou can export lora from sdxl dreambooth
Fi replied you back
EHi, I followed the order, but still clciking the web UI displays "This site can’t be reached 127.0.0.1 refused to connect.". and then again the Ngrok page says not able to start.