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
{{ message }}
Repository navigation
Why is noise scheduler in train_dreambooth.py hard coded to "CompVis/stable-diffusion-v1-4" ? #1264
Is it required for the noise scheduler to be of SD v1.4 or can we use the pretrained_model provided in the command line ?
I want to try the script with v1.5 but not sure if I need to replace the noise scheduler also ?
changed the title [-]Why is noise scheduler in train_dreambooth.py is hard coded to "CompVis/stable-diffusion-v1-4" ? [/-][+]Why is noise scheduler in train_dreambooth.py hard coded to "CompVis/stable-diffusion-v1-4" ? [/+]on Nov 12, 2022
Activity
[-]Why is noise scheduler in train_dreambooth.py is hard coded to "CompVis/stable-diffusion-v1-4" ? [/-][+]Why is noise scheduler in train_dreambooth.py hard coded to "CompVis/stable-diffusion-v1-4" ? [/+]patrickvonplaten commentedon Nov 16, 2022
That's a very good question - it's actually a bug! Thanks for spotting it :-)
Here the PR that should solve it: #1322