From d64b451681bdba5453723d3fe0b0681a470d8045 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Wed, 28 Sep 2022 18:09:06 +0300 Subject: added support for automatically installing latest k-diffusion added eta parameter to parameters output for generated images split eta settings into ancestral and ddim (because they have different default values) --- modules/sd_samplers.py | 83 ++++++++++++++++++++++++++------------------------ 1 file changed, 44 insertions(+), 39 deletions(-) (limited to 'modules/sd_samplers.py') diff --git a/modules/sd_samplers.py b/modules/sd_samplers.py index a1183997..3588aae6 100644 --- a/modules/sd_samplers.py +++ b/modules/sd_samplers.py @@ -40,10 +40,8 @@ samplers_for_img2img = [x for x in samplers if x.name != 'PLMS'] sampler_extra_params = { 'sample_euler': ['s_churn', 's_tmin', 's_tmax', 's_noise'], - 'sample_euler_ancestral': ['eta'], 'sample_heun': ['s_churn', 's_tmin', 's_tmax', 's_noise'], 'sample_dpm_2': ['s_churn', 's_tmin', 's_tmax', 's_noise'], - 'sample_dpm_2_ancestral': ['eta'], } def setup_img2img_steps(p, steps=None): @@ -101,6 +99,8 @@ class VanillaStableDiffusionSampler: self.init_latent = None self.sampler_noises = None self.step = 0 + self.eta = None + self.default_eta = 0.0 def number_of_needed_noises(self, p): return 0 @@ -123,20 +123,29 @@ class VanillaStableDiffusionSampler: self.step += 1 return res + def initialize(self, p): + self.eta = p.eta or opts.eta_ddim + + for fieldname in ['p_sample_ddim', 'p_sample_plms']: + if hasattr(self.sampler, fieldname): + setattr(self.sampler, fieldname, self.p_sample_ddim_hook) + + self.mask = p.mask if hasattr(p, 'mask') else None + self.nmask = p.nmask if hasattr(p, 'nmask') else None + def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None): steps, t_enc = setup_img2img_steps(p, steps) # existing code fails with cetain step counts, like 9 try: - self.sampler.make_schedule(ddim_num_steps=steps, ddim_eta=p.ddim_eta, ddim_discretize=p.ddim_discretize, verbose=False) + self.sampler.make_schedule(ddim_num_steps=steps, ddim_eta=self.eta, ddim_discretize=p.ddim_discretize, verbose=False) except Exception: - self.sampler.make_schedule(ddim_num_steps=steps+1,ddim_eta=p.ddim_eta, ddim_discretize=p.ddim_discretize, verbose=False) + self.sampler.make_schedule(ddim_num_steps=steps+1, ddim_eta=self.eta, ddim_discretize=p.ddim_discretize, verbose=False) x1 = self.sampler.stochastic_encode(x, torch.tensor([t_enc] * int(x.shape[0])).to(shared.device), noise=noise) - self.sampler.p_sample_ddim = self.p_sample_ddim_hook - self.mask = p.mask if hasattr(p, 'mask') else None - self.nmask = p.nmask if hasattr(p, 'nmask') else None + self.initialize(p) + self.init_latent = x self.step = 0 @@ -145,11 +154,8 @@ class VanillaStableDiffusionSampler: return samples def sample(self, p, x, conditioning, unconditional_conditioning, steps=None): - for fieldname in ['p_sample_ddim', 'p_sample_plms']: - if hasattr(self.sampler, fieldname): - setattr(self.sampler, fieldname, self.p_sample_ddim_hook) - self.mask = None - self.nmask = None + self.initialize(p) + self.init_latent = None self.step = 0 @@ -157,9 +163,9 @@ class VanillaStableDiffusionSampler: # existing code fails with cetin step counts, like 9 try: - samples_ddim, _ = self.sampler.sample(S=steps, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=p.eta) + samples_ddim, _ = self.sampler.sample(S=steps, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=self.eta) except Exception: - samples_ddim, _ = self.sampler.sample(S=steps+1, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=p.eta) + samples_ddim, _ = self.sampler.sample(S=steps+1, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=self.eta) return samples_ddim @@ -237,6 +243,8 @@ class KDiffusionSampler: self.sampler_noises = None self.sampler_noise_index = 0 self.stop_at = None + self.eta = None + self.default_eta = 1.0 def callback_state(self, d): store_latent(d["denoised"]) @@ -255,22 +263,12 @@ class KDiffusionSampler: self.sampler_noise_index += 1 return res - def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None): - steps, t_enc = setup_img2img_steps(p, steps) - - sigmas = self.model_wrap.get_sigmas(steps) - - noise = noise * sigmas[steps - t_enc - 1] - - xi = x + noise - - sigma_sched = sigmas[steps - t_enc - 1:] - + def initialize(self, p): self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None - self.model_wrap_cfg.init_latent = x self.model_wrap.step = 0 self.sampler_noise_index = 0 + self.eta = p.eta or opts.eta_ancestral if hasattr(k_diffusion.sampling, 'trange'): k_diffusion.sampling.trange = lambda *args, **kwargs: extended_trange(self, *args, **kwargs) @@ -283,6 +281,25 @@ class KDiffusionSampler: if hasattr(p, param_name) and param_name in inspect.signature(self.func).parameters: extra_params_kwargs[param_name] = getattr(p, param_name) + if 'eta' in inspect.signature(self.func).parameters: + extra_params_kwargs['eta'] = self.eta + + return extra_params_kwargs + + def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None): + steps, t_enc = setup_img2img_steps(p, steps) + + sigmas = self.model_wrap.get_sigmas(steps) + + noise = noise * sigmas[steps - t_enc - 1] + xi = x + noise + + extra_params_kwargs = self.initialize(p) + + sigma_sched = sigmas[steps - t_enc - 1:] + + self.model_wrap_cfg.init_latent = x + return self.func(self.model_wrap_cfg, xi, sigma_sched, extra_args={'cond': conditioning, 'uncond': unconditional_conditioning, 'cond_scale': p.cfg_scale}, disable=False, callback=self.callback_state, **extra_params_kwargs) def sample(self, p, x, conditioning, unconditional_conditioning, steps=None): @@ -291,19 +308,7 @@ class KDiffusionSampler: sigmas = self.model_wrap.get_sigmas(steps) x = x * sigmas[0] - self.model_wrap_cfg.step = 0 - self.sampler_noise_index = 0 - - if hasattr(k_diffusion.sampling, 'trange'): - k_diffusion.sampling.trange = lambda *args, **kwargs: extended_trange(self, *args, **kwargs) - - if self.sampler_noises is not None: - k_diffusion.sampling.torch = TorchHijack(self) - - extra_params_kwargs = {} - for param_name in self.extra_params: - if hasattr(p, param_name) and param_name in inspect.signature(self.func).parameters: - extra_params_kwargs[param_name] = getattr(p, param_name) + extra_params_kwargs = self.initialize(p) samples = self.func(self.model_wrap_cfg, x, sigmas, extra_args={'cond': conditioning, 'uncond': unconditional_conditioning, 'cond_scale': p.cfg_scale}, disable=False, callback=self.callback_state, **extra_params_kwargs) -- cgit v1.2.3