rework Negative Guidance minimum sigma to work with AND, add infotext and... (1d11e896) · Commits · github_fork / Stable Diffusion Webui

javascript/hints.js

+2 −1

Original line number	Original line	Diff line number	Diff line
	@@ -111,7 +111,8 @@ titles = {
	"Resize height to": "Resizes image to this height. If 0, height is inferred from either of two nearby sliders.",		"Resize height to": "Resizes image to this height. If 0, height is inferred from either of two nearby sliders.",
	"Multiplier for extra networks": "When adding extra network such as Hypernetwork or Lora to prompt, use this multiplier for it.",		"Multiplier for extra networks": "When adding extra network such as Hypernetwork or Lora to prompt, use this multiplier for it.",
	"Discard weights with matching name": "Regular expression; if weights's name matches it, the weights is not written to the resulting checkpoint. Use ^model_ema to discard EMA weights.",		"Discard weights with matching name": "Regular expression; if weights's name matches it, the weights is not written to the resulting checkpoint. Use ^model_ema to discard EMA weights.",
	"Extra networks tab order": "Comma-separated list of tab names; tabs listed here will appear in the extra networks UI first and in order lsited."		"Extra networks tab order": "Comma-separated list of tab names; tabs listed here will appear in the extra networks UI first and in order lsited.",
			"Negative Guidance minimum sigma": "Skip negative prompt for steps where image is already mostly denoised; the higher this value, the more skips there will be; provides increased performance in exchange for minor quality reduction."
	}		}

modules/generation_parameters_copypaste.py

+1 −0

Original line number	Original line	Diff line number	Diff line
	@@ -309,6 +309,7 @@ infotext_to_setting_name_mapping = [
	('UniPC order', 'uni_pc_order'),		('UniPC order', 'uni_pc_order'),
	('UniPC lower order final', 'uni_pc_lower_order_final'),		('UniPC lower order final', 'uni_pc_lower_order_final'),
	('RNG', 'randn_source'),		('RNG', 'randn_source'),
			('NGMS', 's_min_uncond'),
	]		]

modules/processing.py

+2 −1

Original line number	Original line	Diff line number	Diff line
	@@ -480,7 +480,8 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter
	"Clip skip": None if clip_skip <= 1 else clip_skip,		"Clip skip": None if clip_skip <= 1 else clip_skip,
	"ENSD": None if opts.eta_noise_seed_delta == 0 else opts.eta_noise_seed_delta,		"ENSD": None if opts.eta_noise_seed_delta == 0 else opts.eta_noise_seed_delta,
	"Init image hash": getattr(p, 'init_img_hash', None),		"Init image hash": getattr(p, 'init_img_hash', None),
	"RNG": (opts.randn_source if opts.randn_source != "GPU" else None)		"RNG": opts.randn_source if opts.randn_source != "GPU" else None,
			"NGMS": None if p.s_min_uncond == 0 else p.s_min_uncond,
	}		}

	generation_params.update(p.extra_generation_params)		generation_params.update(p.extra_generation_params)

modules/sd_samplers_kdiffusion.py

+25 −18

Original line number	Original line	Diff line number	Diff line
	@@ -115,20 +115,21 @@ class CFGDenoiser(torch.nn.Module):
	sigma_in = denoiser_params.sigma		sigma_in = denoiser_params.sigma
	tensor = denoiser_params.text_cond		tensor = denoiser_params.text_cond
	uncond = denoiser_params.text_uncond		uncond = denoiser_params.text_uncond
			skip_uncond = False

	if self.step % 2 and s_min_uncond > 0 and not is_edit_model:
	# alternating uncond allows for higher thresholds without the quality loss normally expected from raising it		# alternating uncond allows for higher thresholds without the quality loss normally expected from raising it
	sigma_threshold = s_min_uncond		if self.step % 2 and s_min_uncond > 0 and sigma[0] < s_min_uncond and not is_edit_model:
	if(torch.dot(sigma,sigma) < sigma.shape[0] * (sigma_threshold*sigma_threshold) ):		skip_uncond = True
	uncond = torch.zeros([0,0,uncond.shape[2]])		x_in = x_in[:-batch_size]
	x_in=x_in[:x_in.shape[0]//2]		sigma_in = sigma_in[:-batch_size]
	sigma_in=sigma_in[:sigma_in.shape[0]//2]

	if tensor.shape[1] == uncond.shape[1]:		if tensor.shape[1] == uncond.shape[1] or skip_uncond:
	if not is_edit_model:		if is_edit_model:
	cond_in = torch.cat([tensor, uncond])
	else:
	cond_in = torch.cat([tensor, uncond, uncond])		cond_in = torch.cat([tensor, uncond, uncond])
			elif skip_uncond:
			cond_in = tensor
			else:
			cond_in = torch.cat([tensor, uncond])

	if shared.batch_cond_uncond:		if shared.batch_cond_uncond:
	x_out = self.inner_model(x_in, sigma_in, cond=make_condition_dict([cond_in], image_cond_in))		x_out = self.inner_model(x_in, sigma_in, cond=make_condition_dict([cond_in], image_cond_in))
	@@ -152,9 +153,15 @@ class CFGDenoiser(torch.nn.Module):

	x_out[a:b] = self.inner_model(x_in[a:b], sigma_in[a:b], cond=make_condition_dict(c_crossattn, image_cond_in[a:b]))		x_out[a:b] = self.inner_model(x_in[a:b], sigma_in[a:b], cond=make_condition_dict(c_crossattn, image_cond_in[a:b]))

	if uncond.shape[0]:		if not skip_uncond:
	x_out[-uncond.shape[0]:] = self.inner_model(x_in[-uncond.shape[0]:], sigma_in[-uncond.shape[0]:], cond=make_condition_dict([uncond], image_cond_in[-uncond.shape[0]:]))		x_out[-uncond.shape[0]:] = self.inner_model(x_in[-uncond.shape[0]:], sigma_in[-uncond.shape[0]:], cond=make_condition_dict([uncond], image_cond_in[-uncond.shape[0]:]))

			if skip_uncond:
			#x_out = torch.cat([x_out, x_out[0:batch_size]]) # we skipped uncond denoising, so we put cond-denoised image to where the uncond-denoised image should be
			denoised_image_indexes = [x[0][0] for x in conds_list]
			fake_uncond = torch.cat([x_out[i:i+1] for i in denoised_image_indexes])
			x_out = torch.cat([x_out, fake_uncond])

	denoised_params = CFGDenoisedParams(x_out, state.sampling_step, state.sampling_steps)		denoised_params = CFGDenoisedParams(x_out, state.sampling_step, state.sampling_steps)
	cfg_denoised_callback(denoised_params)		cfg_denoised_callback(denoised_params)

	@@ -165,13 +172,12 @@ class CFGDenoiser(torch.nn.Module):
	elif opts.live_preview_content == "Negative prompt":		elif opts.live_preview_content == "Negative prompt":
	sd_samplers_common.store_latent(x_out[-uncond.shape[0]:])		sd_samplers_common.store_latent(x_out[-uncond.shape[0]:])

	if not is_edit_model:		if is_edit_model:
	if uncond.shape[0]:
	denoised = self.combine_denoised(x_out, conds_list, uncond, cond_scale)
	else:
	denoised = x_out
	else:
	denoised = self.combine_denoised_for_edit_model(x_out, cond_scale)		denoised = self.combine_denoised_for_edit_model(x_out, cond_scale)
			elif skip_uncond:
			denoised = self.combine_denoised(x_out, conds_list, uncond, 1.0)
			else:
			denoised = self.combine_denoised(x_out, conds_list, uncond, cond_scale)

	if self.mask is not None:		if self.mask is not None:
	denoised = self.init_latent * self.mask + self.nmask * denoised		denoised = self.init_latent * self.mask + self.nmask * denoised
	@@ -221,6 +227,7 @@ class KDiffusionSampler:
	self.eta = None		self.eta = None
	self.config = None		self.config = None
	self.last_latent = None		self.last_latent = None
			self.s_min_uncond = None

	self.conditioning_key = sd_model.model.conditioning_key		self.conditioning_key = sd_model.model.conditioning_key

Admin message