InvokeAI/invokeai/backend
Lincoln Stein 2f9dcd7906 support both epsilon and v-prediction v2 inference
There are actually two Stable Diffusion v2 legacy checkpoint
configurations:

1) "epsilon" prediction type for Stable Diffusion v2 Base
2) "v-prediction" type for Stable Diffusion v2-768

This commit adds the configuration file needed for epsilon prediction
type models as well as the UI that prompts the user to select the
appropriate configuration file when the code can't do so
automatically.
2023-03-05 22:51:40 -05:00
..
config move diagnostic message to stderr; was confusing CI 2023-03-03 01:54:48 -05:00
generator Update txt2img2img.py 2023-03-04 12:58:33 +11:00
image_util all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
model_management support both epsilon and v-prediction v2 inference 2023-03-05 22:51:40 -05:00
prompting all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
restoration all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
stable_diffusion all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
training all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
util all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
web all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
__init__.py fix broken scripts 2023-03-03 20:19:37 -05:00
args.py all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
generate.py all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00
globals.py all vestiges of ldm.invoke removed 2023-03-03 01:02:00 -05:00