OPENGIT/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-04-04 03:29:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix Beta sampling to match the paper

Browse Source 
We need a beta distribution of timesteps, not sigmas. Also allow
distribution parameters > 1.
This commit is contained in:
Jan Alexander Steffens (heftig) 2025-03-28 23:06:26 +01:00
parent 374bb6cc38
commit cdd9964358
No known key found for this signature in database
GPG Key ID: B8AC08600F108CDF
2 changed files with 9 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
modules/sd_schedulers.py
View File

@ -117,12 +117,15 @@ def ddim_scheduler(n, sigma_min, sigma_max, inner_model, device):
def beta_scheduler(n, sigma_min, sigma_max, inner_model, device):
# From "Beta Sampling is All You Need" [arXiv:2407.12173] (Lee et. al, 2024) """
# From "Beta Sampling is All You Need" [arXiv:2407.12173] (Lee et. al, 2024)
alpha = shared.opts.beta_dist_alpha
beta = shared.opts.beta_dist_beta
timesteps = 1 - np.linspace(0, 1, n)
timesteps = [stats.beta.ppf(x, alpha, beta) for x in timesteps]
sigmas = [sigma_min + (x * (sigma_max-sigma_min)) for x in timesteps]
curve = [stats.beta.ppf(x, alpha, beta) for x in np.linspace(1, 0, n)]
start = inner_model.sigma_to_t(torch.tensor(sigma_max))
end = inner_model.sigma_to_t(torch.tensor(sigma_min))
timesteps = [end + x * (start - end) for x in curve]
sigmas = [inner_model.t_to_sigma(ts) for ts in timesteps]
sigmas += [0.0]
return torch.FloatTensor(sigmas).to(device)

4
modules/shared_options.py
View File

@ -407,8 +407,8 @@ options_templates.update(options_section(('sampler-params', "Sampler parameters"
'uni_pc_lower_order_final': OptionInfo(True, "UniPC lower order final", infotext='UniPC lower order final'),
'sd_noise_schedule': OptionInfo("Default", "Noise schedule for sampling", gr.Radio, {"choices": ["Default", "Zero Terminal SNR"]}, infotext="Noise Schedule").info("for use with zero terminal SNR trained models"),
'skip_early_cond': OptionInfo(0.0, "Ignore negative prompt during early sampling", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}, infotext="Skip Early CFG").info("disables CFG on a proportion of steps at the beginning of generation; 0=skip none; 1=skip all; can both improve sample diversity/quality and speed up sampling; XYZ plot: Skip Early CFG"),
'beta_dist_alpha': OptionInfo(0.6, "Beta scheduler - alpha", gr.Slider, {"minimum": 0.01, "maximum": 1.0, "step": 0.01}, infotext='Beta scheduler alpha').info('Default = 0.6; the alpha parameter of the beta distribution used in Beta sampling'),
'beta_dist_beta': OptionInfo(0.6, "Beta scheduler - beta", gr.Slider, {"minimum": 0.01, "maximum": 1.0, "step": 0.01}, infotext='Beta scheduler beta').info('Default = 0.6; the beta parameter of the beta distribution used in Beta sampling'),
'beta_dist_alpha': OptionInfo(0.6, "Beta scheduler - alpha", gr.Slider, {"minimum": 0.01, "maximum": 5.0, "step": 0.01}, infotext='Beta scheduler alpha').info('Default = 0.6; the alpha parameter of the beta distribution used in Beta sampling'),
'beta_dist_beta': OptionInfo(0.6, "Beta scheduler - beta", gr.Slider, {"minimum": 0.01, "maximum": 5.0, "step": 0.01}, infotext='Beta scheduler beta').info('Default = 0.6; the beta parameter of the beta distribution used in Beta sampling'),
}))
options_templates.update(options_section(('postprocessing', "Postprocessing", "postprocessing"), {