You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi there, currently training a realistic SDXL checkpoint with a dataset of 85 000 HD images. Im currently using dreambooth Kohya_ss with an A100 80vram from runpod.
The thing is that the preliminary step where kohya_ss " check the images " taking too long time ( around 12 hours for my whole dataset and this is not even the training of the dataset yet ). :
Rent an a100 is a bit costly so Im wondering if there is a way to speed up this process and if yes, which kohya_ss setting will allow me to do that or at least SAVE this step for the next trainings. Thanks
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hi there, currently training a realistic SDXL checkpoint with a dataset of 85 000 HD images. Im currently using dreambooth Kohya_ss with an A100 80vram from runpod.
The thing is that the preliminary step where kohya_ss " check the images " taking too long time ( around 12 hours for my whole dataset and this is not even the training of the dataset yet ). :
Rent an a100 is a bit costly so Im wondering if there is a way to speed up this process and if yes, which kohya_ss setting will allow me to do that or at least SAVE this step for the next trainings. Thanks
Beta Was this translation helpful? Give feedback.
All reactions