Training 1.5 is a lot faster. first time trying it. 10 minutes vs 1 hour same data set except the c
Training 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.
Rsent you a dm
JHello guys!
I might have missed it, but is there a video of training styles from Dr. Furkan?
I might have missed it, but is there a video of training styles from Dr. Furkan?
RJust wanted to check: you cropper just crops to the aspect ratio and your resized just resizes to the cropped aspect ratio? So you run firt cropper then resizer?
RCropper crops to an aspect ratio and deals aspect ratios only at the current size whereas the resizer resizes only does not crop at all. Is that correct?
DDI have this problem when i extract lora from sdxl dreambooth checkpoints
Fhello record video and dm me
Fi have older video but logic same
FYouTubeSECourses
RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. Get your own #Midjourney level model for free via Style Teaching. 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 #StableDiffusion Tutorials, Au...
https://www.patreon.com/SECoursesPlaylist of #StableDiffusion Tutorials, Au...
Fyes first cropper then resizer
Fresizer resize exact resolution with focusing face
Fadd full fp16
DWhere i need to add full fo16
DFull fp16
JI saw this post on civitai https://civitai.com/articles/3378/which-style-training-model-looking-best-do-you-think
It's done with the same method?
It's done with the same method?
Fi also tried captions in that one
Fso that was the difference
Fmoreover now i am using Patreon SDXL DreamBooth config
JI'll check that out, thank you!
Fyou can use our SOTA captioners on Patreon
FI used them
JI'll check it out, but I wrote captions by hand for all 150 images
JHow many you suggest to use for style?
JImages that is
Fuse as many as possible
Fbut make sure that
Fit doesnt contain repeating subjects
JWill do, all of them are professional photos from a photographer.
JSo quality won't be an issue
Fgreat
ESent, pls let me know.