From 2aec11d263e98580787bc3f3130a09ab2d1cdfc9 Mon Sep 17 00:00:00 2001 From: Elias Sundqvist Date: Fri, 16 Sep 2022 06:40:43 +0000 Subject: Add randomness and denoising strength support to alternative img2img --- scripts/img2imgalt.py | 38 ++++++++++++++++++++++++++++++-------- 1 file changed, 30 insertions(+), 8 deletions(-) (limited to 'scripts/img2imgalt.py') diff --git a/scripts/img2imgalt.py b/scripts/img2imgalt.py index 7813bbcc..6581eaad 100644 --- a/scripts/img2imgalt.py +++ b/scripts/img2imgalt.py @@ -76,10 +76,10 @@ class Script(scripts.Script): original_prompt = gr.Textbox(label="Original prompt", lines=1) cfg = gr.Slider(label="Decode CFG scale", minimum=0.0, maximum=15.0, step=0.1, value=1.0) st = gr.Slider(label="Decode steps", minimum=1, maximum=150, step=1, value=50) + randomness = gr.Slider(label="randomness", minimum=0.0, maximum=1.0, step=0.01, value=0.0) + return [original_prompt, cfg, st, randomness] - return [original_prompt, cfg, st] - - def run(self, p, original_prompt, cfg, st): + def run(self, p, original_prompt, cfg, st, randomness): p.batch_size = 1 p.batch_count = 1 @@ -90,18 +90,40 @@ class Script(scripts.Script): same_everything = same_params and self.cache.latent.shape == lat.shape and np.abs(self.cache.latent-lat).sum() < 100 if same_everything: - noise = self.cache.noise + rec_noise = self.cache.noise else: shared.state.job_count += 1 cond = p.sd_model.get_learned_conditioning(p.batch_size * [original_prompt]) uncond = p.sd_model.get_learned_conditioning(p.batch_size * [""]) - noise = find_noise_for_image(p, cond, uncond, cfg, st) - self.cache = Cached(noise, cfg, st, lat, original_prompt) + rec_noise = find_noise_for_image(p, cond, uncond, cfg, st) + self.cache = Cached(rec_noise, cfg, st, lat, original_prompt) + rand_noise = processing.create_random_tensors(p.init_latent.shape[1:], [p.seed + x + 1 for x in range(p.init_latent.shape[0])]) + + combined_noise = ((1 - randomness) * rec_noise + randomness * rand_noise) / ((randomness**2 + (1-randomness)**2) ** 0.5) + sampler = samplers[p.sampler_index].constructor(p.sd_model) - samples_ddim = sampler.sample(p, noise, conditioning, unconditional_conditioning) - return samples_ddim + sigmas = sampler.model_wrap.get_sigmas(p.steps) + + t_enc = int(min(p.denoising_strength, 0.999) * p.steps) + + noise_dt = combined_noise - ( p.init_latent / sigmas[0] ) + noise_dt = noise_dt * sigmas[p.steps - t_enc - 1] + + noise = p.init_latent + noise_dt + + sigma_sched = sigmas[p.steps - t_enc - 1:] + + sampler.model_wrap_cfg.mask = p.mask + sampler.model_wrap_cfg.nmask = p.nmask + sampler.model_wrap_cfg.init_latent = p.init_latent + + if hasattr(K.sampling, 'trange'): + K.sampling.trange = lambda *args, **kwargs: sd_samplers.extended_trange(*args, **kwargs) + + p.seed = p.seed + 1 + return sampler.func(sampler.model_wrap_cfg, noise, sigma_sched, extra_args={'cond': conditioning, 'uncond': unconditional_conditioning, 'cond_scale': p.cfg_scale}, disable=False, callback=sampler.callback_state) p.sample = sample_extra -- cgit v1.2.1 From a441cd563aa04c8ebfc4d02a4a004f649ad44f71 Mon Sep 17 00:00:00 2001 From: Elias Sundqvist Date: Fri, 16 Sep 2022 06:40:43 +0000 Subject: Reduce code duplication --- scripts/img2imgalt.py | 20 ++++---------------- 1 file changed, 4 insertions(+), 16 deletions(-) (limited to 'scripts/img2imgalt.py') diff --git a/scripts/img2imgalt.py b/scripts/img2imgalt.py index 6581eaad..dbda3255 100644 --- a/scripts/img2imgalt.py +++ b/scripts/img2imgalt.py @@ -105,25 +105,13 @@ class Script(scripts.Script): sampler = samplers[p.sampler_index].constructor(p.sd_model) sigmas = sampler.model_wrap.get_sigmas(p.steps) - - t_enc = int(min(p.denoising_strength, 0.999) * p.steps) noise_dt = combined_noise - ( p.init_latent / sigmas[0] ) - noise_dt = noise_dt * sigmas[p.steps - t_enc - 1] - - noise = p.init_latent + noise_dt - - sigma_sched = sigmas[p.steps - t_enc - 1:] - - sampler.model_wrap_cfg.mask = p.mask - sampler.model_wrap_cfg.nmask = p.nmask - sampler.model_wrap_cfg.init_latent = p.init_latent - - if hasattr(K.sampling, 'trange'): - K.sampling.trange = lambda *args, **kwargs: sd_samplers.extended_trange(*args, **kwargs) - + p.seed = p.seed + 1 - return sampler.func(sampler.model_wrap_cfg, noise, sigma_sched, extra_args={'cond': conditioning, 'uncond': unconditional_conditioning, 'cond_scale': p.cfg_scale}, disable=False, callback=sampler.callback_state) + + return sampler.sample_img2img(p, p.init_latent, noise_dt, conditioning, unconditional_conditioning) + p.sample = sample_extra -- cgit v1.2.1 From b64994b973ad8f4268bf785f25f92b66c8dced40 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Fri, 16 Sep 2022 19:24:48 +0300 Subject: added original negative prompt to img2img alt --- scripts/img2imgalt.py | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'scripts/img2imgalt.py') diff --git a/scripts/img2imgalt.py b/scripts/img2imgalt.py index dbda3255..7f1f53a7 100644 --- a/scripts/img2imgalt.py +++ b/scripts/img2imgalt.py @@ -59,7 +59,7 @@ def find_noise_for_image(p, cond, uncond, cfg_scale, steps): return x / x.std() -Cached = namedtuple("Cached", ["noise", "cfg_scale", "steps", "latent", "original_prompt"]) +Cached = namedtuple("Cached", ["noise", "cfg_scale", "steps", "latent", "original_prompt", "original_negative_prompt"]) class Script(scripts.Script): @@ -74,19 +74,20 @@ class Script(scripts.Script): def ui(self, is_img2img): original_prompt = gr.Textbox(label="Original prompt", lines=1) + original_negative_prompt = gr.Textbox(label="Original negative prompt", lines=1) cfg = gr.Slider(label="Decode CFG scale", minimum=0.0, maximum=15.0, step=0.1, value=1.0) st = gr.Slider(label="Decode steps", minimum=1, maximum=150, step=1, value=50) randomness = gr.Slider(label="randomness", minimum=0.0, maximum=1.0, step=0.01, value=0.0) - return [original_prompt, cfg, st, randomness] + return [original_prompt, original_negative_prompt, cfg, st, randomness] - def run(self, p, original_prompt, cfg, st, randomness): + def run(self, p, original_prompt, original_negative_prompt, cfg, st, randomness): p.batch_size = 1 p.batch_count = 1 def sample_extra(x, conditioning, unconditional_conditioning): lat = (p.init_latent.cpu().numpy() * 10).astype(int) - same_params = self.cache is not None and self.cache.cfg_scale == cfg and self.cache.steps == st and self.cache.original_prompt == original_prompt + same_params = self.cache is not None and self.cache.cfg_scale == cfg and self.cache.steps == st and self.cache.original_prompt == original_prompt and self.cache.original_negative_prompt == original_negative_prompt same_everything = same_params and self.cache.latent.shape == lat.shape and np.abs(self.cache.latent-lat).sum() < 100 if same_everything: @@ -94,9 +95,9 @@ class Script(scripts.Script): else: shared.state.job_count += 1 cond = p.sd_model.get_learned_conditioning(p.batch_size * [original_prompt]) - uncond = p.sd_model.get_learned_conditioning(p.batch_size * [""]) + uncond = p.sd_model.get_learned_conditioning(p.batch_size * [original_negative_prompt]) rec_noise = find_noise_for_image(p, cond, uncond, cfg, st) - self.cache = Cached(rec_noise, cfg, st, lat, original_prompt) + self.cache = Cached(rec_noise, cfg, st, lat, original_prompt, original_negative_prompt) rand_noise = processing.create_random_tensors(p.init_latent.shape[1:], [p.seed + x + 1 for x in range(p.init_latent.shape[0])]) -- cgit v1.2.1