hoelee/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-08-04 11:12:35 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

prompt_parser: allow spaces in schedules, add test, log/ignore errors

Browse Source 
Only build the parser once (at import time) instead of for each step.

doctest is run by simply executing modules/prompt_parser.py
This commit is contained in:
Rae Fu
2022-10-04 09:49:51 -06:00
committed by AUTOMATIC1111
parent 1eb588cbf1
commit 90e911fd54
2 changed files with 95 additions and 54 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
modules/processing.py
View File

@@ -84,7 +84,7 @@ class StableDiffusionProcessing:
self.s_tmin = opts.s_tmin
self.s_tmax = float('inf') # not representable as a standard ui option
self.s_noise = opts.s_noise
if not seed_enable_extras:
self.subseed = -1
self.subseed_strength = 0
@@ -296,7 +296,7 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
assert(len(p.prompt) > 0)
else:
assert p.prompt is not None
devices.torch_gc()
seed = get_fixed_seed(p.seed)
@@ -359,8 +359,8 @@ 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(len(prompts) * [p.negative_prompt], p.steps)
c = prompt_parser.get_learned_conditioning(prompts, p.steps)
uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt], p.steps)
c = prompt_parser.get_learned_conditioning(shared.sd_model, prompts, p.steps)
if len(model_hijack.comments) > 0:
for comment in model_hijack.comments:
@@ -527,7 +527,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
# GC now before running the next img2img to prevent running out of memory
x = None
devices.torch_gc()
samples = self.sampler.sample_img2img(self, samples, noise, conditioning, unconditional_conditioning, steps=self.steps)
return samples