lllyasviel/stable-diffusion-webui-forge
1
0
Fork 0
mirror of https://github.com/lllyasviel/stable-diffusion-webui-forge.git synced 2026-04-30 19:21:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update forge_reference.py

Browse Source
This commit is contained in:
lllyasviel
2024-01-30 18:47:18 -08:00
parent 20f162b84e
commit 4915593e72
1 changed files with 19 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
extensions-builtin/forge_preprocessor_reference/scripts/forge_reference.py
View File

@@ -2,6 +2,7 @@ import torch
from modules_forge.supported_preprocessor import Preprocessor, PreprocessorParameter from modules_forge.supported_preprocessor import Preprocessor, PreprocessorParameter
from modules_forge.shared import add_supported_preprocessor from modules_forge.shared import add_supported_preprocessor
from ldm_patched.modules.samplers import sampling_function
class PreprocessorReference(Preprocessor): class PreprocessorReference(Preprocessor):
@@ -33,6 +34,9 @@ class PreprocessorReference(Preprocessor):
latent_image = vae.encode(cond.movedim(1, -1)) latent_image = vae.encode(cond.movedim(1, -1))
latent_image = process.sd_model.forge_objects.unet.model.latent_format.process_in(latent_image) latent_image = process.sd_model.forge_objects.unet.model.latent_format.process_in(latent_image)
gen_seed = process.seeds[0] + 1
gen_cpu = torch.Generator().manual_seed(gen_seed)
unet = process.sd_model.forge_objects.unet.clone() unet = process.sd_model.forge_objects.unet.clone()
sigma_max = unet.model.model_sampling.percent_to_sigma(start_percent) sigma_max = unet.model.model_sampling.percent_to_sigma(start_percent)
sigma_min = unet.model.model_sampling.percent_to_sigma(end_percent) sigma_min = unet.model.model_sampling.percent_to_sigma(end_percent)
@@ -44,6 +48,9 @@ class PreprocessorReference(Preprocessor):
self.is_recording_style = True self.is_recording_style = True
xt = latent_image.to(x) + torch.randn(x.size(), dtype=x.dtype, generator=gen_cpu).to(x) * sigma
sampling_function(model, xt, timestep, uncond, cond, 1, model_options, seed)
self.is_recording_style = False self.is_recording_style = False
return model, x, timestep, uncond, cond, cond_scale, model_options, seed return model, x, timestep, uncond, cond, cond_scale, model_options, seed
@@ -53,7 +60,10 @@ class PreprocessorReference(Preprocessor):
if not (sigma_min <= sigma <= sigma_max): if not (sigma_min <= sigma <= sigma_max):
return h return h
a = 0 if self.is_recording_style:
a = 0
else:
b = 0
return h return h
@@ -62,7 +72,10 @@ class PreprocessorReference(Preprocessor):
if not (sigma_min <= sigma <= sigma_max): if not (sigma_min <= sigma <= sigma_max):
return q, k, v return q, k, v
a = 0 if self.is_recording_style:
a = 0
else:
b = 0
return q, k, v return q, k, v
@@ -71,7 +84,10 @@ class PreprocessorReference(Preprocessor):
if not (sigma_min <= sigma <= sigma_max): if not (sigma_min <= sigma <= sigma_max):
return h return h
a = 0 if self.is_recording_style:
a = 0
else:
b = 0
return h return h