question about fine tuning the base model

#2
by d222097 - opened

I read your description on the civitai, but there are still some details I don't understand:What script did you use for training (eg. diffusers/ss-kohya)? Do you use dreambooth or directly fine-tune the way of training?It would be highly appreciated that you can give me some directions.

Hi. I used Kohya-ss I installed it following the instructions on the page of this trainer. I used Finetuning :)

Hi. I used Kohya-ss I installed it following the instructions on the page of this trainer. I used Finetuning :)

Thank you for your reply. I'm also very interested in your inpainting model. I'd want to know if the inpainting model is just merged from the Realistic_Vision_V5.1 and base stable-diffusion-inpainting model (not training) or fine-tuned (add 5 addtional channel and training)? Sorry my grammar is poor, I hope I made that clear.

Thank you for your reply. I'm also very interested in your inpainting model. I'd want to know if the inpainting model is just merged from the Realistic_Vision_V5.1 and base stable-diffusion-inpainting model (not training) or fine-tuned (add 5 addtional channel and training)? Sorry my grammar is poor, I hope I made that clear.

Inpainting model is the result of a merge using the "Add Difference" method.
Model A = sd1.5-inpainting
Model B = Your Model
Model C = sd1.5 (7gb)
Multiplier = 1

Thank you for your reply. I'm also very interested in your inpainting model. I'd want to know if the inpainting model is just merged from the Realistic_Vision_V5.1 and base stable-diffusion-inpainting model (not training) or fine-tuned (add 5 addtional channel and training)? Sorry my grammar is poor, I hope I made that clear.

Inpainting model is the result of a merge using the "Add Difference" method.
Model A = sd1.5-inpainting
Model B = Your Model
Model C = sd1.5 (7gb)
Multiplier = 1

Thank you very much for your reply

Sign up or log in to comment