InvokeAI/ldm
Lincoln Stein 3e48b9ff85 cut over from karras to model noise schedule for higher steps
The k_samplers come with a "karras" noise schedule which performs
very well at low step counts but becomes noisy at higher ones.

This commit introduces a threshold (currently 30 steps) at which the
k samplers will switch over from using karras to the older model
noise schedule.
2022-10-22 23:02:50 -04:00
..
data Textual Inversion for M1 2022-09-27 01:39:17 +02:00
invoke add support for loading VAE autoencoders 2022-10-22 19:27:46 -04:00
models cut over from karras to model noise schedule for higher steps 2022-10-22 23:02:50 -04:00
modules Merge branch 'development' into fix-high-step-count 2022-10-21 06:55:17 -04:00
generate.py Merge branch 'development' into patch-1 2022-10-22 19:28:50 -04:00
lr_scheduler.py prettified all the code using "blue" at the urging of @tildebyte 2022-08-26 03:15:42 -04:00
simplet2i.py Squashed commit of the following: 2022-09-12 14:31:48 -04:00
util.py add ability to import and edit alternative models online 2022-10-13 23:48:07 -04:00