Mirror/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2024-06-07 21:20:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #14979 from drhead/refiner_cumprod_fix

Browse Source 
Protect alphas_cumprod during refiner switchover
This commit is contained in:
AUTOMATIC1111 2024-03-02 06:40:32 +03:00
parent 978c7fadb3
commit 28bc85a20a
2 changed files with 33 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
modules/processing.py
View File

@ -915,33 +915,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
if p.n_iter > 1:
shared.state.job = f"Batch {n+1} out of {p.n_iter}"
def rescale_zero_terminal_snr_abar(alphas_cumprod):
alphas_bar_sqrt = alphas_cumprod.sqrt()
# Store old values.
alphas_bar_sqrt_0 = alphas_bar_sqrt[0].clone()
alphas_bar_sqrt_T = alphas_bar_sqrt[-1].clone()
# Shift so the last timestep is zero.
alphas_bar_sqrt -= (alphas_bar_sqrt_T)
# Scale so the first timestep is back to the old value.
alphas_bar_sqrt *= alphas_bar_sqrt_0 / (alphas_bar_sqrt_0 - alphas_bar_sqrt_T)
# Convert alphas_bar_sqrt to betas
alphas_bar = alphas_bar_sqrt**2 # Revert sqrt
alphas_bar[-1] = 4.8973451890853435e-08
return alphas_bar
if hasattr(p.sd_model, 'alphas_cumprod') and hasattr(p.sd_model, 'alphas_cumprod_original'):
p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device)
if opts.use_downcasted_alpha_bar:
p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar
p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device)
if opts.sd_noise_schedule == "Zero Terminal SNR":
p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule
p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device)
sd_models.apply_alpha_schedule_override(p.sd_model, p)
with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast():
samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts)

32
modules/sd_models.py
View File

@ -15,6 +15,7 @@ from ldm.util import instantiate_from_config
from modules import paths, shared, modelloader, devices, script_callbacks, sd_vae, sd_disable_initialization, errors, hashes, sd_models_config, sd_unet, sd_models_xl, cache, extra_networks, processing, lowvram, sd_hijack, patches
from modules.timer import Timer
from modules.shared import opts
import tomesd
import numpy as np
@ -549,6 +550,36 @@ def repair_config(sd_config):
karlo_path = os.path.join(paths.models_path, 'karlo')
sd_config.model.params.noise_aug_config.params.clip_stats_path = sd_config.model.params.noise_aug_config.params.clip_stats_path.replace("checkpoints/karlo_models", karlo_path)
def apply_alpha_schedule_override(sd_model, p=None):
def rescale_zero_terminal_snr_abar(alphas_cumprod):
alphas_bar_sqrt = alphas_cumprod.sqrt()
# Store old values.
alphas_bar_sqrt_0 = alphas_bar_sqrt[0].clone()
alphas_bar_sqrt_T = alphas_bar_sqrt[-1].clone()
# Shift so the last timestep is zero.
alphas_bar_sqrt -= (alphas_bar_sqrt_T)
# Scale so the first timestep is back to the old value.
alphas_bar_sqrt *= alphas_bar_sqrt_0 / (alphas_bar_sqrt_0 - alphas_bar_sqrt_T)
# Convert alphas_bar_sqrt to betas
alphas_bar = alphas_bar_sqrt**2 # Revert sqrt
alphas_bar[-1] = 4.8973451890853435e-08
return alphas_bar
if hasattr(sd_model, 'alphas_cumprod') and hasattr(sd_model, 'alphas_cumprod_original'):
sd_model.alphas_cumprod = sd_model.alphas_cumprod_original.to(shared.device)
if opts.use_downcasted_alpha_bar:
if p is not None:
p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar
sd_model.alphas_cumprod = sd_model.alphas_cumprod.half().to(shared.device)
if opts.sd_noise_schedule == "Zero Terminal SNR":
if p is not None:
p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule
sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(sd_model.alphas_cumprod).to(shared.device)
sd1_clip_weight = 'cond_stage_model.transformer.text_model.embeddings.token_embedding.weight'
sd2_clip_weight = 'cond_stage_model.model.transformer.resblocks.0.attn.in_proj_weight'
@ -812,6 +843,7 @@ def reload_model_weights(sd_model=None, info=None, forced_reload=False):
sd_model = reuse_model_from_already_loaded(sd_model, checkpoint_info, timer)
if not forced_reload and sd_model is not None and sd_model.sd_checkpoint_info.filename == checkpoint_info.filename:
apply_alpha_schedule_override(sd_model)
return sd_model
if sd_model is not None: