gitea-mirror/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-12-21 23:00:59 -08:00
Code Issues Packages Projects Releases Wiki Activity

init

Browse Source
This commit is contained in:
MalumaDev
2022-10-14 10:56:41 +02:00
parent fdecb63685
commit bb57f30c2d
9 changed files with 172 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
modules/processing.py
View File

@@ -316,11 +316,16 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration
return f"{all_prompts[index]}{negative_prompt_text}\n{generation_params_text}".strip()
def process_images(p: StableDiffusionProcessing) -> Processed:
def process_images(p: StableDiffusionProcessing, aesthetic_lr=0, aesthetic_weight=0, aesthetic_steps=0,
aesthetic_imgs=None,aesthetic_slerp=False) -> Processed:
"""this is the main loop that both txt2img and img2img use; it calls func_init once inside all the scopes and func_sample once per batch"""
aesthetic_lr = float(aesthetic_lr)
aesthetic_weight = float(aesthetic_weight)
aesthetic_steps = int(aesthetic_steps)
if type(p.prompt) == list:
assert(len(p.prompt) > 0)
assert (len(p.prompt) > 0)
else:
assert p.prompt is not None
@@ -394,7 +399,13 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
#uc = p.sd_model.get_learned_conditioning(len(prompts) * [p.negative_prompt])
#c = p.sd_model.get_learned_conditioning(prompts)
with devices.autocast():
uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt], p.steps)
if hasattr(shared.sd_model.cond_stage_model, "set_aesthetic_params"):
shared.sd_model.cond_stage_model.set_aesthetic_params(0, 0, 0)
uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt],
p.steps)
if hasattr(shared.sd_model.cond_stage_model, "set_aesthetic_params"):
shared.sd_model.cond_stage_model.set_aesthetic_params(aesthetic_lr, aesthetic_weight,
aesthetic_steps, aesthetic_imgs,aesthetic_slerp)
c = prompt_parser.get_multicond_learned_conditioning(shared.sd_model, prompts, p.steps)
if len(model_hijack.comments) > 0: