From ca45ff1ae6fdd5c2dcd754fde95dd29f49bd414b Mon Sep 17 00:00:00 2001 From: ljleb Date: Mon, 24 Jul 2023 13:52:24 -0400 Subject: add postprocess_batch_list callback --- modules/processing.py | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index a74a5302..c16404f4 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -717,7 +717,25 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.all_subseeds = [int(subseed) + x for x in range(len(p.all_prompts))] def infotext(iteration=0, position_in_batch=0, use_main_prompt=False): - return create_infotext(p, p.all_prompts, p.all_seeds, p.all_subseeds, comments, iteration, position_in_batch, use_main_prompt) + all_prompts = p.all_prompts[:] + all_seeds = p.all_seeds[:] + all_subseeds = p.all_subseeds[:] + + # apply changes to generation data + all_prompts[n * p.batch_size:(n + 1) * p.batch_size] = p.prompts + all_seeds[n * p.batch_size:(n + 1) * p.batch_size] = p.seeds + all_subseeds[n * p.batch_size:(n + 1) * p.batch_size] = p.subseeds + + # update p.all_negative_prompts in case extensions changed the size of the batch + # create_infotext below uses it + old_negative_prompts = p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size] + p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size] = p.negative_prompts + + try: + return create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration, position_in_batch, use_main_prompt) + finally: + # restore p.all_negative_prompts in case extensions changed the size of the batch + p.all_negative_prompts[n * p.batch_size:n * p.batch_size + len(p.negative_prompts)] = old_negative_prompts if os.path.exists(cmd_opts.embeddings_dir) and not p.do_not_reload_embeddings: model_hijack.embedding_db.load_textual_inversion_embeddings() @@ -806,6 +824,10 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.scripts is not None: p.scripts.postprocess_batch(p, x_samples_ddim, batch_number=n) + postprocess_batch_list_args = scripts.PostprocessBatchListArgs(list(x_samples_ddim)) + p.scripts.postprocess_batch_list(p, postprocess_batch_list_args, batch_number=n) + x_samples_ddim = postprocess_batch_list_args.images + for i, x_sample in enumerate(x_samples_ddim): p.batch_index = i -- cgit v1.2.3 From 6b68b590321fcac2ad6d71c5aee1ac02687328d7 Mon Sep 17 00:00:00 2001 From: ljleb Date: Mon, 24 Jul 2023 15:38:52 -0400 Subject: use local vars --- modules/processing.py | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index c16404f4..7043477f 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -722,20 +722,20 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: all_subseeds = p.all_subseeds[:] # apply changes to generation data - all_prompts[n * p.batch_size:(n + 1) * p.batch_size] = p.prompts - all_seeds[n * p.batch_size:(n + 1) * p.batch_size] = p.seeds - all_subseeds[n * p.batch_size:(n + 1) * p.batch_size] = p.subseeds + all_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.prompts + all_seeds[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.seeds + all_subseeds[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.subseeds # update p.all_negative_prompts in case extensions changed the size of the batch # create_infotext below uses it - old_negative_prompts = p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size] - p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size] = p.negative_prompts + old_negative_prompts = p.all_negative_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] + p.all_negative_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.negative_prompts try: return create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration, position_in_batch, use_main_prompt) finally: # restore p.all_negative_prompts in case extensions changed the size of the batch - p.all_negative_prompts[n * p.batch_size:n * p.batch_size + len(p.negative_prompts)] = old_negative_prompts + p.all_negative_prompts[iteration * p.batch_size:iteration * p.batch_size + len(p.negative_prompts)] = old_negative_prompts if os.path.exists(cmd_opts.embeddings_dir) and not p.do_not_reload_embeddings: model_hijack.embedding_db.load_textual_inversion_embeddings() -- cgit v1.2.3 From 5b06607476d1ef2c9d16fe8b21c786b2ca13b95c Mon Sep 17 00:00:00 2001 From: ljleb Date: Mon, 24 Jul 2023 15:43:06 -0400 Subject: simplify --- modules/processing.py | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 7043477f..6dc178e1 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -718,24 +718,26 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: def infotext(iteration=0, position_in_batch=0, use_main_prompt=False): all_prompts = p.all_prompts[:] + all_negative_prompts = p.all_negative_prompts[:] all_seeds = p.all_seeds[:] all_subseeds = p.all_subseeds[:] # apply changes to generation data all_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.prompts + all_negative_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.negative_prompts all_seeds[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.seeds all_subseeds[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.subseeds # update p.all_negative_prompts in case extensions changed the size of the batch # create_infotext below uses it - old_negative_prompts = p.all_negative_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] - p.all_negative_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.negative_prompts + old_negative_prompts = p.all_negative_prompts + p.all_negative_prompts = all_negative_prompts try: return create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration, position_in_batch, use_main_prompt) finally: # restore p.all_negative_prompts in case extensions changed the size of the batch - p.all_negative_prompts[iteration * p.batch_size:iteration * p.batch_size + len(p.negative_prompts)] = old_negative_prompts + p.all_negative_prompts = old_negative_prompts if os.path.exists(cmd_opts.embeddings_dir) and not p.do_not_reload_embeddings: model_hijack.embedding_db.load_textual_inversion_embeddings() -- cgit v1.2.3 From ae36e0899fe912cd701fc4bae5c9d0ce9a5b3e41 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 26 Jul 2023 06:36:06 +0300 Subject: alternative solution for infotext issue --- modules/processing.py | 62 ++++++++++++++++++++++----------------------------- modules/scripts.py | 6 ++++- 2 files changed, 32 insertions(+), 36 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 6dc178e1..146e409a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -600,8 +600,12 @@ def program_version(): return res -def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iteration=0, position_in_batch=0, use_main_prompt=False): - index = position_in_batch + iteration * p.batch_size +def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iteration=0, position_in_batch=0, use_main_prompt=False, index=None, all_negative_prompts=None): + if index is None: + index = position_in_batch + iteration * p.batch_size + + if all_negative_prompts is None: + all_negative_prompts = p.all_negative_prompts clip_skip = getattr(p, 'clip_skip', opts.CLIP_stop_at_last_layers) enable_hr = getattr(p, 'enable_hr', False) @@ -642,7 +646,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter generation_params_text = ", ".join([k if k == v else f'{k}: {generation_parameters_copypaste.quote(v)}' for k, v in generation_params.items() if v is not None]) prompt_text = p.prompt if use_main_prompt else all_prompts[index] - negative_prompt_text = f"\nNegative prompt: {p.all_negative_prompts[index]}" if p.all_negative_prompts[index] else "" + negative_prompt_text = f"\nNegative prompt: {all_negative_prompts[index]}" if all_negative_prompts[index] else "" return f"{prompt_text}{negative_prompt_text}\n{generation_params_text}".strip() @@ -716,29 +720,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: else: p.all_subseeds = [int(subseed) + x for x in range(len(p.all_prompts))] - def infotext(iteration=0, position_in_batch=0, use_main_prompt=False): - all_prompts = p.all_prompts[:] - all_negative_prompts = p.all_negative_prompts[:] - all_seeds = p.all_seeds[:] - all_subseeds = p.all_subseeds[:] - - # apply changes to generation data - all_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.prompts - all_negative_prompts[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.negative_prompts - all_seeds[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.seeds - all_subseeds[iteration * p.batch_size:(iteration + 1) * p.batch_size] = p.subseeds - - # update p.all_negative_prompts in case extensions changed the size of the batch - # create_infotext below uses it - old_negative_prompts = p.all_negative_prompts - p.all_negative_prompts = all_negative_prompts - - try: - return create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration, position_in_batch, use_main_prompt) - finally: - # restore p.all_negative_prompts in case extensions changed the size of the batch - p.all_negative_prompts = old_negative_prompts - if os.path.exists(cmd_opts.embeddings_dir) and not p.do_not_reload_embeddings: model_hijack.embedding_db.load_textual_inversion_embeddings() @@ -826,9 +807,20 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.scripts is not None: p.scripts.postprocess_batch(p, x_samples_ddim, batch_number=n) - postprocess_batch_list_args = scripts.PostprocessBatchListArgs(list(x_samples_ddim)) - p.scripts.postprocess_batch_list(p, postprocess_batch_list_args, batch_number=n) - x_samples_ddim = postprocess_batch_list_args.images + batch_params = scripts.PostprocessBatchListArgs( + list(x_samples_ddim), + p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size], + p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size], + p.seeds, + p.subseeds, + ) + + if p.scripts is not None: + p.scripts.postprocess_batch_list(p, batch_params, batch_number=n) + x_samples_ddim = batch_params.images + + def infotext(index=0, use_main_prompt=False): + return create_infotext(p, batch_params.prompts, batch_params.seeds, batch_params.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=batch_params.negative_prompts) for i, x_sample in enumerate(x_samples_ddim): p.batch_index = i @@ -838,7 +830,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.restore_faces: if opts.save and not p.do_not_save_samples and opts.save_images_before_face_restoration: - images.save_image(Image.fromarray(x_sample), p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(n, i), p=p, suffix="-before-face-restoration") + images.save_image(Image.fromarray(x_sample), p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-face-restoration") devices.torch_gc() @@ -855,15 +847,15 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.color_corrections is not None and i < len(p.color_corrections): if opts.save and not p.do_not_save_samples and opts.save_images_before_color_correction: image_without_cc = apply_overlay(image, p.paste_to, i, p.overlay_images) - images.save_image(image_without_cc, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(n, i), p=p, suffix="-before-color-correction") + images.save_image(image_without_cc, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-color-correction") image = apply_color_correction(p.color_corrections[i], image) image = apply_overlay(image, p.paste_to, i, p.overlay_images) if opts.samples_save and not p.do_not_save_samples: - images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(n, i), p=p) + images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p) - text = infotext(n, i) + text = infotext(i) infotexts.append(text) if opts.enable_pnginfo: image.info["parameters"] = text @@ -874,10 +866,10 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: image_mask_composite = Image.composite(image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') if opts.save_mask: - images.save_image(image_mask, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(n, i), p=p, suffix="-mask") + images.save_image(image_mask, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask") if opts.save_mask_composite: - images.save_image(image_mask_composite, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(n, i), p=p, suffix="-mask-composite") + images.save_image(image_mask_composite, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask-composite") if opts.return_mask: output_images.append(image_mask) diff --git a/modules/scripts.py b/modules/scripts.py index 5b4edcac..1049740d 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -17,8 +17,12 @@ class PostprocessImageArgs: class PostprocessBatchListArgs: - def __init__(self, images): + def __init__(self, images, prompts, negative_prompts, seeds, subseeds): self.images = images + self.prompts = prompts + self.negative_prompts = negative_prompts + self.seeds = seeds + self.subseeds = subseeds class Script: -- cgit v1.2.3 From 7c22bbd3ad5a149e0cf29df887405188fb2d0471 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 26 Jul 2023 07:04:07 +0300 Subject: attempt 2 --- modules/processing.py | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 146e409a..e9108f11 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -621,12 +621,12 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Sampler": p.sampler_name, "CFG scale": p.cfg_scale, "Image CFG scale": getattr(p, 'image_cfg_scale', None), - "Seed": all_seeds[index], + "Seed": p.all_seeds[0] if use_main_prompt else all_seeds[index], "Face restoration": (opts.face_restoration_model if p.restore_faces else None), "Size": f"{p.width}x{p.height}", "Model hash": getattr(p, 'sd_model_hash', None if not opts.add_model_hash_to_info or not shared.sd_model.sd_model_hash else shared.sd_model.sd_model_hash), "Model": (None if not opts.add_model_name_to_info else shared.sd_model.sd_checkpoint_info.name_for_extra), - "Variation seed": (None if p.subseed_strength == 0 else all_subseeds[index]), + "Variation seed": (None if p.subseed_strength == 0 else (p.all_subseeds[0] if use_main_prompt else all_subseeds[index])), "Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength), "Seed resize from": (None if p.seed_resize_from_w <= 0 or p.seed_resize_from_h <= 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"), "Denoising strength": getattr(p, 'denoising_strength', None), @@ -807,20 +807,24 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.scripts is not None: p.scripts.postprocess_batch(p, x_samples_ddim, batch_number=n) - batch_params = scripts.PostprocessBatchListArgs( - list(x_samples_ddim), - p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size], - p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size], - p.seeds, - p.subseeds, - ) + batch_params = scripts.PostprocessBatchListArgs( + list(x_samples_ddim), + p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size], + p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size], + p.seeds, + p.subseeds, + ) - if p.scripts is not None: p.scripts.postprocess_batch_list(p, batch_params, batch_number=n) + x_samples_ddim = batch_params.images + p.prompts = batch_params.prompts + p.negative_prompts = batch_params.negative_prompts + p.seeds = batch_params.seeds + p.subseeds = batch_params.subseeds def infotext(index=0, use_main_prompt=False): - return create_infotext(p, batch_params.prompts, batch_params.seeds, batch_params.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=batch_params.negative_prompts) + return create_infotext(p, p.prompts, p.seeds, p.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=p.negative_prompts) for i, x_sample in enumerate(x_samples_ddim): p.batch_index = i @@ -910,7 +914,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p, images_list=output_images, seed=p.all_seeds[0], - info=infotext(), + info=infotexts[0], comments="".join(f"{comment}\n" for comment in comments), subseed=p.all_subseeds[0], index_of_first_image=index_of_first_image, -- cgit v1.2.3 From 835a7dbf0e73c4cdf945b588d319a6c36652cbe5 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 26 Jul 2023 07:49:57 +0300 Subject: simplify PostprocessBatchListArgs --- modules/processing.py | 15 +++------------ modules/scripts.py | 16 ++++++---------- 2 files changed, 9 insertions(+), 22 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index e9108f11..b0992ee1 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -807,21 +807,12 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.scripts is not None: p.scripts.postprocess_batch(p, x_samples_ddim, batch_number=n) - batch_params = scripts.PostprocessBatchListArgs( - list(x_samples_ddim), - p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size], - p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size], - p.seeds, - p.subseeds, - ) + p.prompts = p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size] + p.negative_prompts = p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size] + batch_params = scripts.PostprocessBatchListArgs(list(x_samples_ddim)) p.scripts.postprocess_batch_list(p, batch_params, batch_number=n) - x_samples_ddim = batch_params.images - p.prompts = batch_params.prompts - p.negative_prompts = batch_params.negative_prompts - p.seeds = batch_params.seeds - p.subseeds = batch_params.subseeds def infotext(index=0, use_main_prompt=False): return create_infotext(p, p.prompts, p.seeds, p.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=p.negative_prompts) diff --git a/modules/scripts.py b/modules/scripts.py index 4317cbb6..5b4edcac 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -17,12 +17,8 @@ class PostprocessImageArgs: class PostprocessBatchListArgs: - def __init__(self, images, prompts, negative_prompts, seeds, subseeds): + def __init__(self, images): self.images = images - self.prompts = prompts - self.negative_prompts = negative_prompts - self.seeds = seeds - self.subseeds = subseeds class Script: @@ -172,11 +168,11 @@ class Script: You can modify the postprocessing object (pp) to update the images in the batch, remove images, add images, etc. If the number of images is different from the batch size when returning, - then the script has the responsibility to also update the following attributes in the processing object (pp): - - pp.prompts - - pp.negative_prompts - - pp.seeds - - pp.subseeds + then the script has the responsibility to also update the following attributes in the processing object (p): + - p.prompts + - p.negative_prompts + - p.seeds + - p.subseeds **kwargs will have same items as process_batch, and also: - batch_number - index of current batch, from 0 to number of batches-1 -- cgit v1.2.3 From 3bca90b249d749ed5429f76e380d2ffa52fc0d41 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 13:48:27 +0300 Subject: hires fix checkpoint selection --- modules/generation_parameters_copypaste.py | 3 ++ modules/processing.py | 47 +++++++++++++++++++----------- modules/sd_models.py | 22 ++++++++------ modules/shared.py | 19 ++++++++---- modules/txt2img.py | 3 +- modules/ui.py | 8 ++++- 6 files changed, 68 insertions(+), 34 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index a3448be9..4e286558 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -280,6 +280,9 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model if "Hires sampler" not in res: res["Hires sampler"] = "Use same sampler" + if "Hires checkpoint" not in res: + res["Hires checkpoint"] = "Use same checkpoint" + if "Hires prompt" not in res: res["Hires prompt"] = "" diff --git a/modules/processing.py b/modules/processing.py index b0992ee1..7026487a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -935,7 +935,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): cached_hr_uc = [None, None] cached_hr_c = [None, None] - def __init__(self, enable_hr: bool = False, denoising_strength: float = 0.75, firstphase_width: int = 0, firstphase_height: int = 0, hr_scale: float = 2.0, hr_upscaler: str = None, hr_second_pass_steps: int = 0, hr_resize_x: int = 0, hr_resize_y: int = 0, hr_sampler_name: str = None, hr_prompt: str = '', hr_negative_prompt: str = '', **kwargs): + def __init__(self, enable_hr: bool = False, denoising_strength: float = 0.75, firstphase_width: int = 0, firstphase_height: int = 0, hr_scale: float = 2.0, hr_upscaler: str = None, hr_second_pass_steps: int = 0, hr_resize_x: int = 0, hr_resize_y: int = 0, hr_checkpoint_name: str = None, hr_sampler_name: str = None, hr_prompt: str = '', hr_negative_prompt: str = '', **kwargs): super().__init__(**kwargs) self.enable_hr = enable_hr self.denoising_strength = denoising_strength @@ -946,11 +946,14 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): self.hr_resize_y = hr_resize_y self.hr_upscale_to_x = hr_resize_x self.hr_upscale_to_y = hr_resize_y + self.hr_checkpoint_name = hr_checkpoint_name + self.hr_checkpoint_info = None self.hr_sampler_name = hr_sampler_name self.hr_prompt = hr_prompt self.hr_negative_prompt = hr_negative_prompt self.all_hr_prompts = None self.all_hr_negative_prompts = None + self.latent_scale_mode = None if firstphase_width != 0 or firstphase_height != 0: self.hr_upscale_to_x = self.width @@ -973,6 +976,14 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def init(self, all_prompts, all_seeds, all_subseeds): if self.enable_hr: + if self.hr_checkpoint_name: + self.hr_checkpoint_info = sd_models.get_closet_checkpoint_match(self.hr_checkpoint_name) + + if self.hr_checkpoint_info is None: + raise Exception(f'Could not find checkpoint with name {self.hr_checkpoint_name}') + + self.extra_generation_params["Hires checkpoint"] = self.hr_checkpoint_info.short_title + if self.hr_sampler_name is not None and self.hr_sampler_name != self.sampler_name: self.extra_generation_params["Hires sampler"] = self.hr_sampler_name @@ -982,6 +993,11 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): if tuple(self.hr_negative_prompt) != tuple(self.negative_prompt): self.extra_generation_params["Hires negative prompt"] = self.hr_negative_prompt + self.latent_scale_mode = shared.latent_upscale_modes.get(self.hr_upscaler, None) if self.hr_upscaler is not None else shared.latent_upscale_modes.get(shared.latent_upscale_default_mode, "nearest") + if self.enable_hr and self.latent_scale_mode is None: + if not any(x.name == self.hr_upscaler for x in shared.sd_upscalers): + raise Exception(f"could not find upscaler named {self.hr_upscaler}") + if opts.use_old_hires_fix_width_height and self.applied_old_hires_behavior_to != (self.width, self.height): self.hr_resize_x = self.width self.hr_resize_y = self.height @@ -1020,14 +1036,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): self.truncate_x = (self.hr_upscale_to_x - target_w) // opt_f self.truncate_y = (self.hr_upscale_to_y - target_h) // opt_f - # special case: the user has chosen to do nothing - if self.hr_upscale_to_x == self.width and self.hr_upscale_to_y == self.height: - self.enable_hr = False - self.denoising_strength = None - self.extra_generation_params.pop("Hires upscale", None) - self.extra_generation_params.pop("Hires resize", None) - return - if not state.processing_has_refined_job_count: if state.job_count == -1: state.job_count = self.n_iter @@ -1045,17 +1053,22 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) - latent_scale_mode = shared.latent_upscale_modes.get(self.hr_upscaler, None) if self.hr_upscaler is not None else shared.latent_upscale_modes.get(shared.latent_upscale_default_mode, "nearest") - if self.enable_hr and latent_scale_mode is None: - if not any(x.name == self.hr_upscaler for x in shared.sd_upscalers): - raise Exception(f"could not find upscaler named {self.hr_upscaler}") - x = create_random_tensors([opt_C, self.height // opt_f, self.width // opt_f], seeds=seeds, subseeds=subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self) samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x)) if not self.enable_hr: return samples + current = shared.sd_model.sd_checkpoint_info + try: + if self.hr_checkpoint_info is not None: + sd_models.reload_model_weights(info=self.hr_checkpoint_info) + + return self.sample_hr_pass(samples, seeds, subseeds, subseed_strength, prompts) + finally: + sd_models.reload_model_weights(info=current) + + def sample_hr_pass(self, samples, seeds, subseeds, subseed_strength, prompts): self.is_hr_pass = True target_width = self.hr_upscale_to_x @@ -1073,11 +1086,11 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): info = create_infotext(self, self.all_prompts, self.all_seeds, self.all_subseeds, [], iteration=self.iteration, position_in_batch=index) images.save_image(image, self.outpath_samples, "", seeds[index], prompts[index], opts.samples_format, info=info, p=self, suffix="-before-highres-fix") - if latent_scale_mode is not None: + if self.latent_scale_mode is not None: for i in range(samples.shape[0]): save_intermediate(samples, i) - samples = torch.nn.functional.interpolate(samples, size=(target_height // opt_f, target_width // opt_f), mode=latent_scale_mode["mode"], antialias=latent_scale_mode["antialias"]) + samples = torch.nn.functional.interpolate(samples, size=(target_height // opt_f, target_width // opt_f), mode=self.latent_scale_mode["mode"], antialias=self.latent_scale_mode["antialias"]) # Avoid making the inpainting conditioning unless necessary as # this does need some extra compute to decode / encode the image again. @@ -1193,7 +1206,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): self.hr_uc = None self.hr_c = None - if self.enable_hr: + if self.enable_hr and self.hr_checkpoint_info is None: if shared.opts.hires_fix_use_firstpass_conds: self.calculate_hr_conds() diff --git a/modules/sd_models.py b/modules/sd_models.py index acb1e817..cb67e425 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -52,6 +52,7 @@ class CheckpointInfo: self.shorthash = self.sha256[0:10] if self.sha256 else None self.title = name if self.shorthash is None else f'{name} [{self.shorthash}]' + self.short_title = self.name_for_extra if self.shorthash is None else f'{self.name_for_extra} [{self.shorthash}]' self.ids = [self.hash, self.model_name, self.title, name, f'{name} [{self.hash}]'] + ([self.shorthash, self.sha256, f'{self.name} [{self.shorthash}]'] if self.shorthash else []) @@ -81,6 +82,7 @@ class CheckpointInfo: checkpoints_list.pop(self.title) self.title = f'{self.name} [{self.shorthash}]' + self.short_title = f'{self.name_for_extra} [{self.shorthash}]' self.register() return self.shorthash @@ -101,14 +103,8 @@ def setup_model(): enable_midas_autodownload() -def checkpoint_tiles(): - def convert(name): - return int(name) if name.isdigit() else name.lower() - - def alphanumeric_key(key): - return [convert(c) for c in re.split('([0-9]+)', key)] - - return sorted([x.title for x in checkpoints_list.values()], key=alphanumeric_key) +def checkpoint_tiles(use_short=False): + return [x.short_title if use_short else x.title for x in checkpoints_list.values()] def list_models(): @@ -131,11 +127,14 @@ def list_models(): elif cmd_ckpt is not None and cmd_ckpt != shared.default_sd_model_file: print(f"Checkpoint in --ckpt argument not found (Possible it was moved to {model_path}: {cmd_ckpt}", file=sys.stderr) - for filename in sorted(model_list, key=str.lower): + for filename in model_list: checkpoint_info = CheckpointInfo(filename) checkpoint_info.register() +re_strip_checksum = re.compile(r"\s*\[[^]]+]\s*$") + + def get_closet_checkpoint_match(search_string): checkpoint_info = checkpoint_aliases.get(search_string, None) if checkpoint_info is not None: @@ -145,6 +144,11 @@ def get_closet_checkpoint_match(search_string): if found: return found[0] + search_string_without_checksum = re.sub(re_strip_checksum, '', search_string) + found = sorted([info for info in checkpoints_list.values() if search_string_without_checksum in info.title], key=lambda x: len(x.title)) + if found: + return found[0] + return None diff --git a/modules/shared.py b/modules/shared.py index aa72c9c8..807fb9e3 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -220,12 +220,19 @@ class State: return import modules.sd_samplers - if opts.show_progress_grid: - self.assign_current_image(modules.sd_samplers.samples_to_image_grid(self.current_latent)) - else: - self.assign_current_image(modules.sd_samplers.sample_to_image(self.current_latent)) - self.current_image_sampling_step = self.sampling_step + try: + if opts.show_progress_grid: + self.assign_current_image(modules.sd_samplers.samples_to_image_grid(self.current_latent)) + else: + self.assign_current_image(modules.sd_samplers.sample_to_image(self.current_latent)) + + self.current_image_sampling_step = self.sampling_step + + except Exception: + # when switching models during genration, VAE would be on CPU, so creating an image will fail. + # we silently ignore this error + errors.record_exception() def assign_current_image(self, image): self.current_image = image @@ -512,7 +519,7 @@ options_templates.update(options_section(('ui', "User interface"), { "ui_tab_order": OptionInfo([], "UI tab order", ui_components.DropdownMulti, lambda: {"choices": list(tab_names)}).needs_restart(), "hidden_tabs": OptionInfo([], "Hidden UI tabs", ui_components.DropdownMulti, lambda: {"choices": list(tab_names)}).needs_restart(), "ui_reorder_list": OptionInfo([], "txt2img/img2img UI item order", ui_components.DropdownMulti, lambda: {"choices": list(shared_items.ui_reorder_categories())}).info("selected items appear first").needs_restart(), - "hires_fix_show_sampler": OptionInfo(False, "Hires fix: show hires sampler selection").needs_restart(), + "hires_fix_show_sampler": OptionInfo(False, "Hires fix: show hires checkpoint and sampler selection").needs_restart(), "hires_fix_show_prompts": OptionInfo(False, "Hires fix: show hires prompt and negative prompt").needs_restart(), "disable_token_counters": OptionInfo(False, "Disable prompt token counters").needs_restart(), })) diff --git a/modules/txt2img.py b/modules/txt2img.py index 29d94e8c..935ed418 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -9,7 +9,7 @@ from modules.ui import plaintext_to_html import gradio as gr -def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_sampler_index: int, hr_prompt: str, hr_negative_prompt, override_settings_texts, request: gr.Request, *args): +def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_index: int, hr_prompt: str, hr_negative_prompt, override_settings_texts, request: gr.Request, *args): override_settings = create_override_settings_dict(override_settings_texts) p = processing.StableDiffusionProcessingTxt2Img( @@ -41,6 +41,7 @@ def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, step hr_second_pass_steps=hr_second_pass_steps, hr_resize_x=hr_resize_x, hr_resize_y=hr_resize_y, + hr_checkpoint_name=None if hr_checkpoint_name == 'Use same checkpoint' else hr_checkpoint_name, hr_sampler_name=sd_samplers.samplers_for_img2img[hr_sampler_index - 1].name if hr_sampler_index != 0 else None, hr_prompt=hr_prompt, hr_negative_prompt=hr_negative_prompt, diff --git a/modules/ui.py b/modules/ui.py index 07ecee7b..6d8265f2 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -476,6 +476,10 @@ def create_ui(): hr_resize_y = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize height to", value=0, elem_id="txt2img_hr_resize_y") with FormRow(elem_id="txt2img_hires_fix_row3", variant="compact", visible=opts.hires_fix_show_sampler) as hr_sampler_container: + checkpoint_choices = lambda: ["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True) + hr_checkpoint_name = gr.Dropdown(label='Hires checkpoint', elem_id="hr_checkpoint", choices=checkpoint_choices(), value="Use same checkpoint") + create_refresh_button(hr_checkpoint_name, modules.sd_models.list_models, lambda: {"choices": checkpoint_choices()}, "hr_checkpoint_refresh") + hr_sampler_index = gr.Dropdown(label='Hires sampling method', elem_id="hr_sampler", choices=["Use same sampler"] + [x.name for x in samplers_for_img2img], value="Use same sampler", type="index") with FormRow(elem_id="txt2img_hires_fix_row4", variant="compact", visible=opts.hires_fix_show_prompts) as hr_prompts_container: @@ -553,6 +557,7 @@ def create_ui(): hr_second_pass_steps, hr_resize_x, hr_resize_y, + hr_checkpoint_name, hr_sampler_index, hr_prompt, hr_negative_prompt, @@ -630,8 +635,9 @@ def create_ui(): (hr_second_pass_steps, "Hires steps"), (hr_resize_x, "Hires resize-1"), (hr_resize_y, "Hires resize-2"), + (hr_checkpoint_name, "Hires checkpoint"), (hr_sampler_index, "Hires sampler"), - (hr_sampler_container, lambda d: gr.update(visible=True) if d.get("Hires sampler", "Use same sampler") != "Use same sampler" else gr.update()), + (hr_sampler_container, lambda d: gr.update(visible=True) if d.get("Hires sampler", "Use same sampler") != "Use same sampler" or d.get("Hires checkpoint", "Use same checkpoint") != "Use same checkpoint" else gr.update()), (hr_prompt, "Hires prompt"), (hr_negative_prompt, "Hires negative prompt"), (hr_prompts_container, lambda d: gr.update(visible=True) if d.get("Hires prompt", "") != "" or d.get("Hires negative prompt", "") != "" else gr.update()), -- cgit v1.2.3 From 40cd59207b96f9e522fdc104b43279880b671ce4 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 14:10:26 +0300 Subject: make it work with SDXL --- modules/processing.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 7026487a..b8af1301 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1197,8 +1197,11 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): if self.hr_c is not None: return - self.hr_uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, self.hr_negative_prompts, self.steps * self.step_multiplier, [self.cached_hr_uc, self.cached_uc], self.hr_extra_network_data) - self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, self.hr_prompts, self.steps * self.step_multiplier, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) + hr_prompts = prompt_parser.SdConditioning(self.hr_prompts, width=self.hr_upscale_to_x, height=self.hr_upscale_to_y) + hr_negative_prompts = prompt_parser.SdConditioning(self.hr_negative_prompts, width=self.hr_upscale_to_x, height=self.hr_upscale_to_y, is_negative_prompt=True) + + self.hr_uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, hr_negative_prompts, self.steps * self.step_multiplier, [self.cached_hr_uc, self.cached_uc], self.hr_extra_network_data) + self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, hr_prompts, self.steps * self.step_multiplier, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) def setup_conds(self): super().setup_conds() -- cgit v1.2.3 From 77761e7bad8a7cbffc9028dc0b2f63169aaf25f9 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 14:10:33 +0300 Subject: linter --- modules/processing.py | 2 +- modules/ui.py | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b8af1301..21dbef16 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1055,6 +1055,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): x = create_random_tensors([opt_C, self.height // opt_f, self.width // opt_f], seeds=seeds, subseeds=subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self) samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x)) + del x if not self.enable_hr: return samples @@ -1137,7 +1138,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): noise = create_random_tensors(samples.shape[1:], seeds=seeds, subseeds=subseeds, subseed_strength=subseed_strength, p=self) # GC now before running the next img2img to prevent running out of memory - x = None devices.torch_gc() if not self.disable_extra_networks: diff --git a/modules/ui.py b/modules/ui.py index 6d8265f2..6fc9de83 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -476,9 +476,9 @@ def create_ui(): hr_resize_y = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize height to", value=0, elem_id="txt2img_hr_resize_y") with FormRow(elem_id="txt2img_hires_fix_row3", variant="compact", visible=opts.hires_fix_show_sampler) as hr_sampler_container: - checkpoint_choices = lambda: ["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True) - hr_checkpoint_name = gr.Dropdown(label='Hires checkpoint', elem_id="hr_checkpoint", choices=checkpoint_choices(), value="Use same checkpoint") - create_refresh_button(hr_checkpoint_name, modules.sd_models.list_models, lambda: {"choices": checkpoint_choices()}, "hr_checkpoint_refresh") + + hr_checkpoint_name = gr.Dropdown(label='Hires checkpoint', elem_id="hr_checkpoint", choices=["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True), value="Use same checkpoint") + create_refresh_button(hr_checkpoint_name, modules.sd_models.list_models, lambda: {"choices": ["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True)}, "hr_checkpoint_refresh") hr_sampler_index = gr.Dropdown(label='Hires sampling method', elem_id="hr_sampler", choices=["Use same sampler"] + [x.name for x in samplers_for_img2img], value="Use same sampler", type="index") -- cgit v1.2.3 From eec540b22798ddcf8a03d947519c36635d77d722 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 15:04:12 +0300 Subject: repair non-latent upscaling broken for SDXL --- modules/processing.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 21dbef16..6fb14516 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1119,6 +1119,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = torch.from_numpy(np.array(batch_images)) decoded_samples = decoded_samples.to(shared.device) decoded_samples = 2. * decoded_samples - 1. + decoded_samples = decoded_samples.to(shared.device, dtype=devices.dtype_vae) samples = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(decoded_samples)) -- cgit v1.2.3 From a64fbe89288802f8b5ec8ca7bcab5aaf2c7bfea5 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 15:12:09 +0300 Subject: make it possible to use checkpoints of different types (SD1, SDXL) in first and second pass of hires fix --- modules/processing.py | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 6fb14516..c4da208f 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1060,16 +1060,21 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): if not self.enable_hr: return samples + if self.latent_scale_mode is None: + decoded_samples = decode_first_stage(self.sd_model, samples) + else: + decoded_samples = None + current = shared.sd_model.sd_checkpoint_info try: if self.hr_checkpoint_info is not None: sd_models.reload_model_weights(info=self.hr_checkpoint_info) - return self.sample_hr_pass(samples, seeds, subseeds, subseed_strength, prompts) + return self.sample_hr_pass(samples, decoded_samples, seeds, subseeds, subseed_strength, prompts) finally: sd_models.reload_model_weights(info=current) - def sample_hr_pass(self, samples, seeds, subseeds, subseed_strength, prompts): + def sample_hr_pass(self, samples, decoded_samples, seeds, subseeds, subseed_strength, prompts): self.is_hr_pass = True target_width = self.hr_upscale_to_x @@ -1100,7 +1105,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): else: image_conditioning = self.txt2img_image_conditioning(samples) else: - decoded_samples = decode_first_stage(self.sd_model, samples) lowres_samples = torch.clamp((decoded_samples + 1.0) / 2.0, min=0.0, max=1.0) batch_images = [] -- cgit v1.2.3 From cc53db6652b11e6f7bca42c3aa93bd6761ed3d3f Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 15:30:33 +0300 Subject: this time for sure --- modules/processing.py | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index c4da208f..3190b964 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -538,8 +538,12 @@ def create_random_tensors(shape, seeds, subseeds=None, subseed_strength=0.0, see return x +class DecodedSamples(list): + already_decoded = True + + def decode_latent_batch(model, batch, target_device=None, check_for_nans=False): - samples = [] + samples = DecodedSamples() for i in range(batch.shape[0]): sample = decode_first_stage(model, batch[i:i + 1])[0] @@ -793,7 +797,11 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: 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) - x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) + if getattr(samples_ddim, 'already_decoded', False): + x_samples_ddim = samples_ddim + else: + x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) + x_samples_ddim = torch.stack(x_samples_ddim).float() x_samples_ddim = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0) @@ -1161,9 +1169,11 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): sd_models.apply_token_merging(self.sd_model, self.get_token_merging_ratio()) + decoded_samples = decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True) + self.is_hr_pass = False - return samples + return decoded_samples def close(self): super().close() -- cgit v1.2.3 From 02038036ff571e0f04a94c3e279609666e239dec Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 16:16:31 +0300 Subject: make it so that VAE NaNs autodetection also works during first pass of hires fix --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 3190b964..0677de81 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1069,7 +1069,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): return samples if self.latent_scale_mode is None: - decoded_samples = decode_first_stage(self.sd_model, samples) + decoded_samples = torch.stack(decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)) else: decoded_samples = None -- cgit v1.2.3 From 0af4127fd14360ebb12c6569d98aebf8047abbfc Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 30 Jul 2023 19:36:24 +0300 Subject: delete the field that is preventing the model from being unloaded and is causing increased RAM usage --- modules/processing.py | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 0677de81..b09433b0 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1076,11 +1076,15 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): current = shared.sd_model.sd_checkpoint_info try: if self.hr_checkpoint_info is not None: + del self.sampler sd_models.reload_model_weights(info=self.hr_checkpoint_info) + devices.torch_gc() return self.sample_hr_pass(samples, decoded_samples, seeds, subseeds, subseed_strength, prompts) finally: + del self.sampler sd_models.reload_model_weights(info=current) + devices.torch_gc() def sample_hr_pass(self, samples, decoded_samples, seeds, subseeds, subseed_strength, prompts): self.is_hr_pass = True -- cgit v1.2.3 From dca121e9035ba36b3f7484c8a31a7776d85c0960 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 31 Jul 2023 09:13:07 +0300 Subject: set the field to None instead --- modules/processing.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b09433b0..35e7b87e 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1076,13 +1076,13 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): current = shared.sd_model.sd_checkpoint_info try: if self.hr_checkpoint_info is not None: - del self.sampler + self.sampler = None sd_models.reload_model_weights(info=self.hr_checkpoint_info) devices.torch_gc() return self.sample_hr_pass(samples, decoded_samples, seeds, subseeds, subseed_strength, prompts) finally: - del self.sampler + self.sampler = None sd_models.reload_model_weights(info=current) devices.torch_gc() -- cgit v1.2.3 From 29d7e31d89e9d686784eacbdbfc5b15959eb4449 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 31 Jul 2023 10:43:26 +0300 Subject: repair AttributeError: 'NoneType' object has no attribute 'conditioning_key' --- modules/processing.py | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 35e7b87e..1f0c0b3b 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1104,6 +1104,13 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): info = create_infotext(self, self.all_prompts, self.all_seeds, self.all_subseeds, [], iteration=self.iteration, position_in_batch=index) images.save_image(image, self.outpath_samples, "", seeds[index], prompts[index], opts.samples_format, info=info, p=self, suffix="-before-highres-fix") + img2img_sampler_name = self.hr_sampler_name or self.sampler_name + + if self.sampler_name in ['PLMS', 'UniPC']: # PLMS/UniPC do not support img2img so we just silently switch to DDIM + img2img_sampler_name = 'DDIM' + + self.sampler = sd_samplers.create_sampler(img2img_sampler_name, self.sd_model) + if self.latent_scale_mode is not None: for i in range(samples.shape[0]): save_intermediate(samples, i) @@ -1143,13 +1150,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): shared.state.nextjob() - img2img_sampler_name = self.hr_sampler_name or self.sampler_name - - if self.sampler_name in ['PLMS', 'UniPC']: # PLMS/UniPC do not support img2img so we just silently switch to DDIM - img2img_sampler_name = 'DDIM' - - self.sampler = sd_samplers.create_sampler(img2img_sampler_name, self.sd_model) - samples = samples[:, :, self.truncate_y//2:samples.shape[2]-(self.truncate_y+1)//2, self.truncate_x//2:samples.shape[3]-(self.truncate_x+1)//2] noise = create_random_tensors(samples.shape[1:], seeds=seeds, subseeds=subseeds, subseed_strength=subseed_strength, p=self) -- cgit v1.2.3 From c09bc2c60856ca1ab2243386176badf909affdbe Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 31 Jul 2023 13:20:26 +0300 Subject: fix "clamp_scalar_cpu" not implemented for 'Half' --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 1f0c0b3b..f8f8bddc 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1069,7 +1069,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): return samples if self.latent_scale_mode is None: - decoded_samples = torch.stack(decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)) + decoded_samples = torch.stack(decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)).to(dtype=torch.float32) else: decoded_samples = None -- cgit v1.2.3 From ccb92339348f6973de39cde062982a51a4cd0818 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 2 Aug 2023 18:53:09 +0300 Subject: add yet another torch_gc to reclaim some of VRAM after the initial stage of img2img --- modules/processing.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b0992ee1..0b66cd2a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1348,6 +1348,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image = image.to(shared.device, dtype=devices.dtype_vae) self.init_latent = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(image)) + devices.torch_gc() if self.resize_mode == 3: self.init_latent = torch.nn.functional.interpolate(self.init_latent, size=(self.height // opt_f, self.width // opt_f), mode="bilinear") -- cgit v1.2.3 From 84b6fcd02ca6d6ab48c4b6be4bb8724b1c2e7014 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 3 Aug 2023 00:00:23 +0300 Subject: add NV option for Random number generator source setting, which allows to generate same pictures on CPU/AMD/Mac as on NVidia videocards. --- modules/devices.py | 39 ++++++++++++++- modules/processing.py | 6 +-- modules/rng_philox.py | 100 ++++++++++++++++++++++++++++++++++++++ modules/sd_samplers_kdiffusion.py | 5 +- modules/shared.py | 2 +- 5 files changed, 142 insertions(+), 10 deletions(-) create mode 100644 modules/rng_philox.py (limited to 'modules/processing.py') diff --git a/modules/devices.py b/modules/devices.py index 57e51da3..b58776d8 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -3,7 +3,7 @@ import contextlib from functools import lru_cache import torch -from modules import errors +from modules import errors, rng_philox if sys.platform == "darwin": from modules import mac_specific @@ -90,23 +90,58 @@ def cond_cast_float(input): return input.float() if unet_needs_upcast else input +nv_rng = None + + def randn(seed, shape): from modules.shared import opts - torch.manual_seed(seed) + manual_seed(seed) + + if opts.randn_source == "NV": + return torch.asarray(nv_rng.randn(shape), device=device) + if opts.randn_source == "CPU" or device.type == 'mps': return torch.randn(shape, device=cpu).to(device) + return torch.randn(shape, device=device) +def randn_like(x): + from modules.shared import opts + + if opts.randn_source == "NV": + return torch.asarray(nv_rng.randn(x.shape), device=x.device, dtype=x.dtype) + + if opts.randn_source == "CPU" or x.device.type == 'mps': + return torch.randn_like(x, device=cpu).to(x.device) + + return torch.randn_like(x) + + def randn_without_seed(shape): from modules.shared import opts + if opts.randn_source == "NV": + return torch.asarray(nv_rng.randn(shape), device=device) + if opts.randn_source == "CPU" or device.type == 'mps': return torch.randn(shape, device=cpu).to(device) + return torch.randn(shape, device=device) +def manual_seed(seed): + from modules.shared import opts + + if opts.randn_source == "NV": + global nv_rng + nv_rng = rng_philox.Generator(seed) + return + + torch.manual_seed(seed) + + def autocast(disable=False): from modules import shared diff --git a/modules/processing.py b/modules/processing.py index 0b66cd2a..8f34c8b4 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -492,7 +492,7 @@ def create_random_tensors(shape, seeds, subseeds=None, subseed_strength=0.0, see noise_shape = shape if seed_resize_from_h <= 0 or seed_resize_from_w <= 0 else (shape[0], seed_resize_from_h//8, seed_resize_from_w//8) subnoise = None - if subseeds is not None: + if subseeds is not None and subseed_strength != 0: subseed = 0 if i >= len(subseeds) else subseeds[i] subnoise = devices.randn(subseed, noise_shape) @@ -524,7 +524,7 @@ def create_random_tensors(shape, seeds, subseeds=None, subseed_strength=0.0, see cnt = p.sampler.number_of_needed_noises(p) if eta_noise_seed_delta > 0: - torch.manual_seed(seed + eta_noise_seed_delta) + devices.manual_seed(seed + eta_noise_seed_delta) for j in range(cnt): sampler_noises[j].append(devices.randn_without_seed(tuple(noise_shape))) @@ -636,7 +636,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Token merging ratio": None if token_merging_ratio == 0 else token_merging_ratio, "Token merging ratio hr": None if not enable_hr or token_merging_ratio_hr == 0 else token_merging_ratio_hr, "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" and opts.randn_source != "NV" else None, "NGMS": None if p.s_min_uncond == 0 else p.s_min_uncond, **p.extra_generation_params, "Version": program_version() if opts.add_version_to_infotext else None, diff --git a/modules/rng_philox.py b/modules/rng_philox.py new file mode 100644 index 00000000..b5c02483 --- /dev/null +++ b/modules/rng_philox.py @@ -0,0 +1,100 @@ +"""RNG imitiating torch cuda randn on CPU. You are welcome. + +Usage: + +``` +g = Generator(seed=0) +print(g.randn(shape=(3, 4))) +``` + +Expected output: +``` +[[-0.92466259 -0.42534415 -2.6438457 0.14518388] + [-0.12086647 -0.57972564 -0.62285122 -0.32838709] + [-1.07454231 -0.36314407 -1.67105067 2.26550497]] +``` +""" + +import numpy as np + +philox_m = [0xD2511F53, 0xCD9E8D57] +philox_w = [0x9E3779B9, 0xBB67AE85] + +two_pow32_inv = np.array([2.3283064e-10], dtype=np.float32) +two_pow32_inv_2pi = np.array([2.3283064e-10 * 6.2831855], dtype=np.float32) + + +def uint32(x): + """Converts (N,) np.uint64 array into (2, N) np.unit32 array.""" + return np.moveaxis(x.view(np.uint32).reshape(-1, 2), 0, 1) + + +def philox4_round(counter, key): + """A single round of the Philox 4x32 random number generator.""" + + v1 = uint32(counter[0].astype(np.uint64) * philox_m[0]) + v2 = uint32(counter[2].astype(np.uint64) * philox_m[1]) + + counter[0] = v2[1] ^ counter[1] ^ key[0] + counter[1] = v2[0] + counter[2] = v1[1] ^ counter[3] ^ key[1] + counter[3] = v1[0] + + +def philox4_32(counter, key, rounds=10): + """Generates 32-bit random numbers using the Philox 4x32 random number generator. + + Parameters: + counter (numpy.ndarray): A 4xN array of 32-bit integers representing the counter values (offset into generation). + key (numpy.ndarray): A 2xN array of 32-bit integers representing the key values (seed). + rounds (int): The number of rounds to perform. + + Returns: + numpy.ndarray: A 4xN array of 32-bit integers containing the generated random numbers. + """ + + for _ in range(rounds - 1): + philox4_round(counter, key) + + key[0] = key[0] + philox_w[0] + key[1] = key[1] + philox_w[1] + + philox4_round(counter, key) + return counter + + +def box_muller(x, y): + """Returns just the first out of two numbers generated by Box–Muller transform algorithm.""" + u = x.astype(np.float32) * two_pow32_inv + two_pow32_inv / 2 + v = y.astype(np.float32) * two_pow32_inv_2pi + two_pow32_inv_2pi / 2 + + s = np.sqrt(-2.0 * np.log(u)) + + r1 = s * np.sin(v) + return r1.astype(np.float32) + + +class Generator: + """RNG that produces same outputs as torch.randn(..., device='cuda') on CPU""" + + def __init__(self, seed): + self.seed = seed + self.offset = 0 + + def randn(self, shape): + """Generate a sequence of n standard normal random variables using the Philox 4x32 random number generator and the Box-Muller transform.""" + + n = 1 + for x in shape: + n *= x + + counter = np.zeros((4, n), dtype=np.uint32) + counter[0] = self.offset + counter[2] = np.arange(n, dtype=np.uint32) # up to 2^32 numbers can be generated - if you want more you'd need to spill into counter[3] + self.offset += 1 + + key = uint32(np.array([[self.seed] * n], dtype=np.uint64)) + + g = philox4_32(counter, key) + + return box_muller(g[0], g[1]).reshape(shape) # discard g[2] and g[3] diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index e0da3425..d72c1b5f 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -260,10 +260,7 @@ class TorchHijack: if noise.shape == x.shape: return noise - if opts.randn_source == "CPU" or x.device.type == 'mps': - return torch.randn_like(x, device=devices.cpu).to(x.device) - else: - return torch.randn_like(x) + return devices.randn_like(x) class KDiffusionSampler: diff --git a/modules/shared.py b/modules/shared.py index aa72c9c8..7103b4ca 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -428,7 +428,7 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "CLIP_stop_at_last_layers": OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}).link("wiki", "https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#clip-skip").info("ignore last layers of CLIP network; 1 ignores none, 2 ignores one layer"), "upcast_attn": OptionInfo(False, "Upcast cross attention layer to float32"), "auto_vae_precision": OptionInfo(True, "Automaticlly revert VAE to 32-bit floats").info("triggers when a tensor with NaNs is produced in VAE; disabling the option in this case will result in a black square image"), - "randn_source": OptionInfo("GPU", "Random number generator source.", gr.Radio, {"choices": ["GPU", "CPU"]}).info("changes seeds drastically; use CPU to produce the same picture across different videocard vendors"), + "randn_source": OptionInfo("GPU", "Random number generator source.", gr.Radio, {"choices": ["GPU", "CPU", "NV"]}).info("changes seeds drastically; use CPU to produce the same picture across different videocard vendors; use NV to produce same picture as on NVidia videocards"), })) options_templates.update(options_section(('sdxl', "Stable Diffusion XL"), { -- cgit v1.2.3 From a1825ee741bb21b35561d58db8cb316d7f5d0c79 Mon Sep 17 00:00:00 2001 From: Splendide Imaginarius <119545140+Splendide-Imaginarius@users.noreply.github.com> Date: Thu, 3 Aug 2023 02:03:35 +0000 Subject: Make StableDiffusionProcessingImg2Img.mask_blur a property MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes breakage when mask_blur is set after construction. See https://github.com/Coyote-A/ultimate-upscale-for-automatic1111/issues/111#issuecomment-1652091424 Thanks to Алексей Трофимов and eunnone for reporting the issue. --- modules/processing.py | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) mode change 100644 => 100755 modules/processing.py (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py old mode 100644 new mode 100755 index b0992ee1..44d20fb7 --- a/modules/processing.py +++ b/modules/processing.py @@ -1232,11 +1232,10 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.image_mask = mask self.latent_mask = None self.mask_for_overlay = None - if mask_blur is not None: - mask_blur_x = mask_blur - mask_blur_y = mask_blur self.mask_blur_x = mask_blur_x self.mask_blur_y = mask_blur_y + if mask_blur is not None: + self.mask_blur = mask_blur self.inpainting_fill = inpainting_fill self.inpaint_full_res = inpaint_full_res self.inpaint_full_res_padding = inpaint_full_res_padding @@ -1246,6 +1245,22 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.nmask = None self.image_conditioning = None + @property + def mask_blur(self): + if self.mask_blur_x == self.mask_blur_y: + return self.mask_blur_x + return None + + @mask_blur.setter + def mask_blur(self, value): + self.mask_blur_x = value + self.mask_blur_y = value + + @mask_blur.deleter + def mask_blur(self): + del self.mask_blur_x + del self.mask_blur_y + def init(self, all_prompts, all_seeds, all_subseeds): self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) crop_region = None -- cgit v1.2.3 From 75336dfc84cae280036bc52a6805eb10d9ae30ba Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Fri, 4 Aug 2023 13:38:52 +0800 Subject: add TAESD for i2i and t2i --- modules/processing.py | 13 +++++------ modules/sd_samplers_common.py | 38 ++++++++++++++++++++++++++----- modules/sd_vae_approx.py | 2 +- modules/sd_vae_taesd.py | 52 ++++++++++++++++++++++++++++++++++++------- modules/shared.py | 2 ++ 5 files changed, 86 insertions(+), 21 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 8f34c8b4..099d86b7 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -573,9 +573,10 @@ def decode_latent_batch(model, batch, target_device=None, check_for_nans=False): def decode_first_stage(model, x): - x = model.decode_first_stage(x.to(devices.dtype_vae)) - - return x + from modules.sd_samplers_common import samples_to_images_tensor, approximation_indexes + x = x.to(devices.dtype_vae) + approx_index = approximation_indexes.get(opts.sd_vae_decode_method, 0) + return samples_to_images_tensor(x, approx_index, model) def get_fixed_seed(seed): @@ -1344,10 +1345,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less") image = torch.from_numpy(batch_images) - image = 2. * image - 1. - image = image.to(shared.device, dtype=devices.dtype_vae) - - self.init_latent = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(image)) + from modules.sd_samplers_common import images_tensor_to_samples, approximation_indexes + self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() if self.resize_mode == 3: diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 5deda761..5a45e8eb 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -23,19 +23,29 @@ def setup_img2img_steps(p, steps=None): approximation_indexes = {"Full": 0, "Approx NN": 1, "Approx cheap": 2, "TAESD": 3} -def single_sample_to_image(sample, approximation=None): +def samples_to_images_tensor(sample, approximation=None, model=None): + '''latents -> images [-1, 1]''' if approximation is None: approximation = approximation_indexes.get(opts.show_progress_type, 0) if approximation == 2: - x_sample = sd_vae_approx.cheap_approximation(sample) * 0.5 + 0.5 + x_sample = sd_vae_approx.cheap_approximation(sample) elif approximation == 1: - x_sample = sd_vae_approx.model()(sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach() * 0.5 + 0.5 + x_sample = sd_vae_approx.model()(sample.to(devices.device, devices.dtype)).detach() elif approximation == 3: x_sample = sample * 1.5 - x_sample = sd_vae_taesd.model()(x_sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach() + x_sample = sd_vae_taesd.decoder_model()(x_sample.to(devices.device, devices.dtype)).detach() + x_sample = x_sample * 2 - 1 else: - x_sample = processing.decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0] * 0.5 + 0.5 + if model is None: + model = shared.sd_model + x_sample = model.decode_first_stage(sample) + + return x_sample + + +def single_sample_to_image(sample, approximation=None): + x_sample = samples_to_images_tensor(sample.unsqueeze(0), approximation)[0] * 0.5 + 0.5 x_sample = torch.clamp(x_sample, min=0.0, max=1.0) x_sample = 255. * np.moveaxis(x_sample.cpu().numpy(), 0, 2) @@ -52,6 +62,24 @@ def samples_to_image_grid(samples, approximation=None): return images.image_grid([single_sample_to_image(sample, approximation) for sample in samples]) +def images_tensor_to_samples(image, approximation=None, model=None): + '''image[0, 1] -> latent''' + if approximation is None: + approximation = approximation_indexes.get(opts.sd_vae_encode_method, 0) + + if approximation == 3: + image = image.to(devices.device, devices.dtype) + x_latent = sd_vae_taesd.encoder_model()(image) / 1.5 + else: + if model is None: + model = shared.sd_model + image = image.to(shared.device, dtype=devices.dtype_vae) + image = image * 2 - 1 + x_latent = model.get_first_stage_encoding(model.encode_first_stage(image)) + + return x_latent + + def store_latent(decoded): state.current_latent = decoded diff --git a/modules/sd_vae_approx.py b/modules/sd_vae_approx.py index 86bd658a..3965e223 100644 --- a/modules/sd_vae_approx.py +++ b/modules/sd_vae_approx.py @@ -81,6 +81,6 @@ def cheap_approximation(sample): coefs = torch.tensor(coeffs).to(sample.device) - x_sample = torch.einsum("lxy,lr -> rxy", sample, coefs) + x_sample = torch.einsum("...lxy,lr -> ...rxy", sample, coefs) return x_sample diff --git a/modules/sd_vae_taesd.py b/modules/sd_vae_taesd.py index 5bf7c76e..808eb362 100644 --- a/modules/sd_vae_taesd.py +++ b/modules/sd_vae_taesd.py @@ -44,7 +44,17 @@ def decoder(): ) -class TAESD(nn.Module): +def encoder(): + return nn.Sequential( + conv(3, 64), Block(64, 64), + conv(64, 64, stride=2, bias=False), Block(64, 64), Block(64, 64), Block(64, 64), + conv(64, 64, stride=2, bias=False), Block(64, 64), Block(64, 64), Block(64, 64), + conv(64, 64, stride=2, bias=False), Block(64, 64), Block(64, 64), Block(64, 64), + conv(64, 4), + ) + + +class TAESDDecoder(nn.Module): latent_magnitude = 3 latent_shift = 0.5 @@ -55,21 +65,28 @@ class TAESD(nn.Module): self.decoder.load_state_dict( torch.load(decoder_path, map_location='cpu' if devices.device.type != 'cuda' else None)) - @staticmethod - def unscale_latents(x): - """[0, 1] -> raw latents""" - return x.sub(TAESD.latent_shift).mul(2 * TAESD.latent_magnitude) + +class TAESDEncoder(nn.Module): + latent_magnitude = 3 + latent_shift = 0.5 + + def __init__(self, encoder_path="taesd_encoder.pth"): + """Initialize pretrained TAESD on the given device from the given checkpoints.""" + super().__init__() + self.encoder = encoder() + self.encoder.load_state_dict( + torch.load(encoder_path, map_location='cpu' if devices.device.type != 'cuda' else None)) def download_model(model_path, model_url): if not os.path.exists(model_path): os.makedirs(os.path.dirname(model_path), exist_ok=True) - print(f'Downloading TAESD decoder to: {model_path}') + print(f'Downloading TAESD model to: {model_path}') torch.hub.download_url_to_file(model_url, model_path) -def model(): +def decoder_model(): model_name = "taesdxl_decoder.pth" if getattr(shared.sd_model, 'is_sdxl', False) else "taesd_decoder.pth" loaded_model = sd_vae_taesd_models.get(model_name) @@ -78,7 +95,7 @@ def model(): download_model(model_path, 'https://github.com/madebyollin/taesd/raw/main/' + model_name) if os.path.exists(model_path): - loaded_model = TAESD(model_path) + loaded_model = TAESDDecoder(model_path) loaded_model.eval() loaded_model.to(devices.device, devices.dtype) sd_vae_taesd_models[model_name] = loaded_model @@ -86,3 +103,22 @@ def model(): raise FileNotFoundError('TAESD model not found') return loaded_model.decoder + + +def encoder_model(): + model_name = "taesdxl_encoder.pth" if getattr(shared.sd_model, 'is_sdxl', False) else "taesd_encoder.pth" + loaded_model = sd_vae_taesd_models.get(model_name) + + if loaded_model is None: + model_path = os.path.join(paths_internal.models_path, "VAE-taesd", model_name) + download_model(model_path, 'https://github.com/madebyollin/taesd/raw/main/' + model_name) + + if os.path.exists(model_path): + loaded_model = TAESDEncoder(model_path) + loaded_model.eval() + loaded_model.to(devices.device, devices.dtype) + sd_vae_taesd_models[model_name] = loaded_model + else: + raise FileNotFoundError('TAESD model not found') + + return loaded_model.encoder diff --git a/modules/shared.py b/modules/shared.py index cec030f7..61ba9347 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -430,6 +430,8 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "upcast_attn": OptionInfo(False, "Upcast cross attention layer to float32"), "auto_vae_precision": OptionInfo(True, "Automaticlly revert VAE to 32-bit floats").info("triggers when a tensor with NaNs is produced in VAE; disabling the option in this case will result in a black square image"), "randn_source": OptionInfo("GPU", "Random number generator source.", gr.Radio, {"choices": ["GPU", "CPU", "NV"]}).info("changes seeds drastically; use CPU to produce the same picture across different videocard vendors; use NV to produce same picture as on NVidia videocards"), + "sd_vae_encode_method": OptionInfo("Full", "VAE type for encode", gr.Radio, {"choices": ["Full", "TAESD"]}).info("method to encode image to latent (use in img2img or inpaint mask)"), + "sd_vae_decode_method": OptionInfo("Full", "VAE type for decode", gr.Radio, {"choices": ["Full", "TAESD"]}).info("method to decode latent to image"), })) options_templates.update(options_section(('sdxl', "Stable Diffusion XL"), { -- cgit v1.2.3 From f0c1063a707a4a43823b0ed00e2a8eeb22a9ed0a Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Fri, 4 Aug 2023 09:09:09 +0300 Subject: resolve some of circular import issues for kohaku --- modules/hypernetworks/hypernetwork.py | 5 ++--- modules/processing.py | 7 +------ modules/sd_hijack.py | 6 +++--- modules/sd_samplers_common.py | 10 ++++++++-- modules/textual_inversion/textual_inversion.py | 4 +++- 5 files changed, 17 insertions(+), 15 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/hypernetworks/hypernetwork.py b/modules/hypernetworks/hypernetwork.py index c4821d21..70f1cbd2 100644 --- a/modules/hypernetworks/hypernetwork.py +++ b/modules/hypernetworks/hypernetwork.py @@ -10,7 +10,7 @@ import torch import tqdm from einops import rearrange, repeat from ldm.util import default -from modules import devices, processing, sd_models, shared, sd_samplers, hashes, sd_hijack_checkpoint, errors +from modules import devices, sd_models, shared, sd_samplers, hashes, sd_hijack_checkpoint, errors from modules.textual_inversion import textual_inversion, logging from modules.textual_inversion.learn_schedule import LearnRateScheduler from torch import einsum @@ -469,8 +469,7 @@ def create_hypernetwork(name, enable_sizes, overwrite_old, layer_structure=None, def train_hypernetwork(id_task, hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, varsize, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, use_weight, create_image_every, save_hypernetwork_every, template_filename, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height): - # images allows training previews to have infotext. Importing it at the top causes a circular import problem. - from modules import images + from modules import images, processing save_hypernetwork_every = save_hypernetwork_every or 0 create_image_every = create_image_every or 0 diff --git a/modules/processing.py b/modules/processing.py index 8f34c8b4..8086a2b0 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -30,6 +30,7 @@ from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion from einops import repeat, rearrange from blendmodes.blend import blendLayers, BlendType +decode_first_stage = sd_samplers_common.decode_first_stage # some of those options should not be changed at all because they would break the model, so I removed them from options. opt_C = 4 @@ -572,12 +573,6 @@ def decode_latent_batch(model, batch, target_device=None, check_for_nans=False): return samples -def decode_first_stage(model, x): - x = model.decode_first_stage(x.to(devices.dtype_vae)) - - return x - - def get_fixed_seed(seed): if seed is None or seed == '' or seed == -1: return int(random.randrange(4294967294)) diff --git a/modules/sd_hijack.py b/modules/sd_hijack.py index cfa5f0eb..609fd56c 100644 --- a/modules/sd_hijack.py +++ b/modules/sd_hijack.py @@ -2,7 +2,6 @@ import torch from torch.nn.functional import silu from types import MethodType -import modules.textual_inversion.textual_inversion from modules import devices, sd_hijack_optimizations, shared, script_callbacks, errors, sd_unet from modules.hypernetworks import hypernetwork from modules.shared import cmd_opts @@ -164,12 +163,13 @@ class StableDiffusionModelHijack: clip = None optimization_method = None - embedding_db = modules.textual_inversion.textual_inversion.EmbeddingDatabase() - def __init__(self): + import modules.textual_inversion.textual_inversion + self.extra_generation_params = {} self.comments = [] + self.embedding_db = modules.textual_inversion.textual_inversion.EmbeddingDatabase() self.embedding_db.add_embedding_dir(cmd_opts.embeddings_dir) def apply_optimizations(self, option=None): diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 5deda761..b3d344e7 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -2,7 +2,7 @@ from collections import namedtuple import numpy as np import torch from PIL import Image -from modules import devices, processing, images, sd_vae_approx, sd_samplers, sd_vae_taesd, shared +from modules import devices, images, sd_vae_approx, sd_samplers, sd_vae_taesd, shared from modules.shared import opts, state SamplerData = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options']) @@ -35,7 +35,7 @@ def single_sample_to_image(sample, approximation=None): x_sample = sample * 1.5 x_sample = sd_vae_taesd.model()(x_sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach() else: - x_sample = processing.decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0] * 0.5 + 0.5 + x_sample = decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0] * 0.5 + 0.5 x_sample = torch.clamp(x_sample, min=0.0, max=1.0) x_sample = 255. * np.moveaxis(x_sample.cpu().numpy(), 0, 2) @@ -44,6 +44,12 @@ def single_sample_to_image(sample, approximation=None): return Image.fromarray(x_sample) +def decode_first_stage(model, x): + x = model.decode_first_stage(x.to(devices.dtype_vae)) + + return x + + def sample_to_image(samples, index=0, approximation=None): return single_sample_to_image(samples[index], approximation) diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index 4713bc2d..aa79dc09 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -13,7 +13,7 @@ import numpy as np from PIL import Image, PngImagePlugin from torch.utils.tensorboard import SummaryWriter -from modules import shared, devices, sd_hijack, processing, sd_models, images, sd_samplers, sd_hijack_checkpoint, errors, hashes +from modules import shared, devices, sd_hijack, sd_models, images, sd_samplers, sd_hijack_checkpoint, errors, hashes import modules.textual_inversion.dataset from modules.textual_inversion.learn_schedule import LearnRateScheduler @@ -387,6 +387,8 @@ def validate_train_inputs(model_name, learn_rate, batch_size, gradient_step, dat def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, varsize, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, use_weight, create_image_every, save_embedding_every, template_filename, save_image_with_stored_embedding, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height): + from modules import processing + save_embedding_every = save_embedding_every or 0 create_image_every = create_image_every or 0 template_file = textual_inversion_templates.get(template_filename, None) -- cgit v1.2.3 From 094c416a801b16c7d8e1944e2e9fae2c9e98bf12 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Fri, 4 Aug 2023 17:53:16 +0800 Subject: change all encode --- modules/processing.py | 14 ++++++-------- modules/sd_samplers_common.py | 2 +- run.ps1 | 1 + run_local.ps1 | 3 +++ update.ps1 | 1 + 5 files changed, 12 insertions(+), 9 deletions(-) create mode 100644 run.ps1 create mode 100644 run_local.ps1 create mode 100644 update.ps1 (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index aae39866..544667a4 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -16,6 +16,7 @@ from typing import Any, Dict, List import modules.sd_hijack from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors from modules.sd_hijack import model_hijack +from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes from modules.shared import opts, cmd_opts, state import modules.shared as shared import modules.paths as paths @@ -30,7 +31,6 @@ from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion from einops import repeat, rearrange from blendmodes.blend import blendLayers, BlendType -decode_first_stage = sd_samplers_common.decode_first_stage # some of those options should not be changed at all because they would break the model, so I removed them from options. opt_C = 4 @@ -84,7 +84,7 @@ def txt2img_image_conditioning(sd_model, x, width, height): # The "masked-image" in this case will just be all zeros since the entire image is masked. image_conditioning = torch.zeros(x.shape[0], 3, height, width, device=x.device) - image_conditioning = sd_model.get_first_stage_encoding(sd_model.encode_first_stage(image_conditioning)) + image_conditioning = images_tensor_to_samples(image_conditioning, approximation_indexes.get(opts.sd_vae_encode_method)) # Add the fake full 1s mask to the first dimension. image_conditioning = torch.nn.functional.pad(image_conditioning, (0, 0, 0, 0, 1, 0), value=1.0) @@ -203,7 +203,7 @@ class StableDiffusionProcessing: midas_in = torch.from_numpy(transformed["midas_in"][None, ...]).to(device=shared.device) midas_in = repeat(midas_in, "1 ... -> n ...", n=self.batch_size) - conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(source_image)) + conditioning_image = images_tensor_to_samples(source_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) conditioning = torch.nn.functional.interpolate( self.sd_model.depth_model(midas_in), size=conditioning_image.shape[2:], @@ -216,7 +216,7 @@ class StableDiffusionProcessing: return conditioning def edit_image_conditioning(self, source_image): - conditioning_image = self.sd_model.encode_first_stage(source_image).mode() + conditioning_image = images_tensor_to_samples(source_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) return conditioning_image @@ -255,7 +255,7 @@ class StableDiffusionProcessing: ) # Encode the new masked image using first stage of network. - conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(conditioning_image)) + conditioning_image = images_tensor_to_samples(conditioning_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) # Create the concatenated conditioning tensor to be fed to `c_concat` conditioning_mask = torch.nn.functional.interpolate(conditioning_mask, size=latent_image.shape[-2:]) @@ -1099,9 +1099,8 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = torch.from_numpy(np.array(batch_images)) decoded_samples = decoded_samples.to(shared.device) - decoded_samples = 2. * decoded_samples - 1. - samples = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(decoded_samples)) + samples = images_tensor_to_samples(decoded_samples, approximation_indexes.get(opts.sd_vae_encode_method)) image_conditioning = self.img2img_image_conditioning(decoded_samples, samples) @@ -1339,7 +1338,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less") image = torch.from_numpy(batch_images) - from modules.sd_samplers_common import images_tensor_to_samples, approximation_indexes self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 7269514f..42a29fc9 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -75,7 +75,7 @@ def images_tensor_to_samples(image, approximation=None, model=None): if approximation == 3: image = image.to(devices.device, devices.dtype) - x_latent = sd_vae_taesd.encoder_model()(image) / 1.5 + x_latent = sd_vae_taesd.encoder_model()(image) else: if model is None: model = shared.sd_model diff --git a/run.ps1 b/run.ps1 new file mode 100644 index 00000000..82c1660b --- /dev/null +++ b/run.ps1 @@ -0,0 +1 @@ +.\venv\Scripts\accelerate-launch.exe --num_cpu_threads_per_process=6 --api .\launch.py --listen --port 17415 --xformers --opt-channelslast \ No newline at end of file diff --git a/run_local.ps1 b/run_local.ps1 new file mode 100644 index 00000000..e2ac43db --- /dev/null +++ b/run_local.ps1 @@ -0,0 +1,3 @@ +.\venv\Scripts\Activate.ps1 +python .\launch.py --xformers --opt-channelslast --api +. $PSCommandPath \ No newline at end of file diff --git a/update.ps1 b/update.ps1 new file mode 100644 index 00000000..9960bead --- /dev/null +++ b/update.ps1 @@ -0,0 +1 @@ +git stash push && git pull --rebase && git stash pop \ No newline at end of file -- cgit v1.2.3 From 6346d8eeaa17ba0f7e41618908519f6e9bfe07e0 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Fri, 4 Aug 2023 17:53:30 +0800 Subject: Revert "change all encode" This reverts commit 094c416a801b16c7d8e1944e2e9fae2c9e98bf12. --- modules/processing.py | 14 ++++++++------ modules/sd_samplers_common.py | 2 +- run.ps1 | 1 - run_local.ps1 | 3 --- update.ps1 | 1 - 5 files changed, 9 insertions(+), 12 deletions(-) delete mode 100644 run.ps1 delete mode 100644 run_local.ps1 delete mode 100644 update.ps1 (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 544667a4..aae39866 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -16,7 +16,6 @@ from typing import Any, Dict, List import modules.sd_hijack from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors from modules.sd_hijack import model_hijack -from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes from modules.shared import opts, cmd_opts, state import modules.shared as shared import modules.paths as paths @@ -31,6 +30,7 @@ from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion from einops import repeat, rearrange from blendmodes.blend import blendLayers, BlendType +decode_first_stage = sd_samplers_common.decode_first_stage # some of those options should not be changed at all because they would break the model, so I removed them from options. opt_C = 4 @@ -84,7 +84,7 @@ def txt2img_image_conditioning(sd_model, x, width, height): # The "masked-image" in this case will just be all zeros since the entire image is masked. image_conditioning = torch.zeros(x.shape[0], 3, height, width, device=x.device) - image_conditioning = images_tensor_to_samples(image_conditioning, approximation_indexes.get(opts.sd_vae_encode_method)) + image_conditioning = sd_model.get_first_stage_encoding(sd_model.encode_first_stage(image_conditioning)) # Add the fake full 1s mask to the first dimension. image_conditioning = torch.nn.functional.pad(image_conditioning, (0, 0, 0, 0, 1, 0), value=1.0) @@ -203,7 +203,7 @@ class StableDiffusionProcessing: midas_in = torch.from_numpy(transformed["midas_in"][None, ...]).to(device=shared.device) midas_in = repeat(midas_in, "1 ... -> n ...", n=self.batch_size) - conditioning_image = images_tensor_to_samples(source_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) + conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(source_image)) conditioning = torch.nn.functional.interpolate( self.sd_model.depth_model(midas_in), size=conditioning_image.shape[2:], @@ -216,7 +216,7 @@ class StableDiffusionProcessing: return conditioning def edit_image_conditioning(self, source_image): - conditioning_image = images_tensor_to_samples(source_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) + conditioning_image = self.sd_model.encode_first_stage(source_image).mode() return conditioning_image @@ -255,7 +255,7 @@ class StableDiffusionProcessing: ) # Encode the new masked image using first stage of network. - conditioning_image = images_tensor_to_samples(conditioning_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) + conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(conditioning_image)) # Create the concatenated conditioning tensor to be fed to `c_concat` conditioning_mask = torch.nn.functional.interpolate(conditioning_mask, size=latent_image.shape[-2:]) @@ -1099,8 +1099,9 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = torch.from_numpy(np.array(batch_images)) decoded_samples = decoded_samples.to(shared.device) + decoded_samples = 2. * decoded_samples - 1. - samples = images_tensor_to_samples(decoded_samples, approximation_indexes.get(opts.sd_vae_encode_method)) + samples = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(decoded_samples)) image_conditioning = self.img2img_image_conditioning(decoded_samples, samples) @@ -1338,6 +1339,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less") image = torch.from_numpy(batch_images) + from modules.sd_samplers_common import images_tensor_to_samples, approximation_indexes self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 42a29fc9..7269514f 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -75,7 +75,7 @@ def images_tensor_to_samples(image, approximation=None, model=None): if approximation == 3: image = image.to(devices.device, devices.dtype) - x_latent = sd_vae_taesd.encoder_model()(image) + x_latent = sd_vae_taesd.encoder_model()(image) / 1.5 else: if model is None: model = shared.sd_model diff --git a/run.ps1 b/run.ps1 deleted file mode 100644 index 82c1660b..00000000 --- a/run.ps1 +++ /dev/null @@ -1 +0,0 @@ -.\venv\Scripts\accelerate-launch.exe --num_cpu_threads_per_process=6 --api .\launch.py --listen --port 17415 --xformers --opt-channelslast \ No newline at end of file diff --git a/run_local.ps1 b/run_local.ps1 deleted file mode 100644 index e2ac43db..00000000 --- a/run_local.ps1 +++ /dev/null @@ -1,3 +0,0 @@ -.\venv\Scripts\Activate.ps1 -python .\launch.py --xformers --opt-channelslast --api -. $PSCommandPath \ No newline at end of file diff --git a/update.ps1 b/update.ps1 deleted file mode 100644 index 9960bead..00000000 --- a/update.ps1 +++ /dev/null @@ -1 +0,0 @@ -git stash push && git pull --rebase && git stash pop \ No newline at end of file -- cgit v1.2.3 From 21000f13a169263a7da2c66b300130d7a6339c7d Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Fri, 4 Aug 2023 18:23:14 +0800 Subject: replace get_first_stage_encoding --- modules/processing.py | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index aae39866..aa6d4d2a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -16,6 +16,7 @@ from typing import Any, Dict, List import modules.sd_hijack from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors from modules.sd_hijack import model_hijack +from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes from modules.shared import opts, cmd_opts, state import modules.shared as shared import modules.paths as paths @@ -30,7 +31,6 @@ from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion from einops import repeat, rearrange from blendmodes.blend import blendLayers, BlendType -decode_first_stage = sd_samplers_common.decode_first_stage # some of those options should not be changed at all because they would break the model, so I removed them from options. opt_C = 4 @@ -84,7 +84,7 @@ def txt2img_image_conditioning(sd_model, x, width, height): # The "masked-image" in this case will just be all zeros since the entire image is masked. image_conditioning = torch.zeros(x.shape[0], 3, height, width, device=x.device) - image_conditioning = sd_model.get_first_stage_encoding(sd_model.encode_first_stage(image_conditioning)) + image_conditioning = images_tensor_to_samples(image_conditioning, approximation_indexes.get(opts.sd_vae_encode_method)) # Add the fake full 1s mask to the first dimension. image_conditioning = torch.nn.functional.pad(image_conditioning, (0, 0, 0, 0, 1, 0), value=1.0) @@ -203,7 +203,7 @@ class StableDiffusionProcessing: midas_in = torch.from_numpy(transformed["midas_in"][None, ...]).to(device=shared.device) midas_in = repeat(midas_in, "1 ... -> n ...", n=self.batch_size) - conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(source_image)) + conditioning_image = images_tensor_to_samples(source_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) conditioning = torch.nn.functional.interpolate( self.sd_model.depth_model(midas_in), size=conditioning_image.shape[2:], @@ -216,7 +216,7 @@ class StableDiffusionProcessing: return conditioning def edit_image_conditioning(self, source_image): - conditioning_image = self.sd_model.encode_first_stage(source_image).mode() + conditioning_image = images_tensor_to_samples(source_image*0.5+0.5, approximation_indexes.get(opts.sd_vae_encode_method)) return conditioning_image @@ -1099,9 +1099,8 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = torch.from_numpy(np.array(batch_images)) decoded_samples = decoded_samples.to(shared.device) - decoded_samples = 2. * decoded_samples - 1. - samples = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(decoded_samples)) + samples = images_tensor_to_samples(decoded_samples, approximation_indexes.get(opts.sd_vae_encode_method)) image_conditioning = self.img2img_image_conditioning(decoded_samples, samples) @@ -1339,7 +1338,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less") image = torch.from_numpy(batch_images) - from modules.sd_samplers_common import images_tensor_to_samples, approximation_indexes self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() -- cgit v1.2.3 From aa744cadc8e357e696a608c8d0c77a7bfc1c9f39 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 5 Aug 2023 12:35:40 +0800 Subject: add infotext --- modules/generation_parameters_copypaste.py | 8 ++++++++ modules/processing.py | 3 +++ 2 files changed, 11 insertions(+) (limited to 'modules/processing.py') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index a3448be9..0713dbf0 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -304,6 +304,12 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model if "Schedule rho" not in res: res["Schedule rho"] = 0 + if "VAE Encoder" not in res: + res["VAE Encoder"] = "Full" + + if "VAE Decoder" not in res: + res["VAE Decoder"] = "Full" + return res @@ -329,6 +335,8 @@ infotext_to_setting_name_mapping = [ ('RNG', 'randn_source'), ('NGMS', 's_min_uncond'), ('Pad conds', 'pad_cond_uncond'), + ('VAE Encoder', 'sd_vae_encode_method'), + ('VAE Decoder', 'sd_vae_decode_method'), ] diff --git a/modules/processing.py b/modules/processing.py index aa6d4d2a..a9ee7507 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -788,6 +788,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: 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) + p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) x_samples_ddim = torch.stack(x_samples_ddim).float() x_samples_ddim = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0) @@ -1100,6 +1101,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = torch.from_numpy(np.array(batch_images)) decoded_samples = decoded_samples.to(shared.device) + self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method samples = images_tensor_to_samples(decoded_samples, approximation_indexes.get(opts.sd_vae_encode_method)) image_conditioning = self.img2img_image_conditioning(decoded_samples, samples) @@ -1338,6 +1340,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less") image = torch.from_numpy(batch_images) + self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() -- cgit v1.2.3 From 56888644a67298253260eda84ceb2d6cd0ce5099 Mon Sep 17 00:00:00 2001 From: Splendide Imaginarius <119545140+Splendide-Imaginarius@users.noreply.github.com> Date: Sat, 5 Aug 2023 04:54:23 +0000 Subject: Reduce mask blur kernel size to 2.5 sigmas MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This more closely matches the old behavior of PIL's Gaussian blur, and fixes breakage when tiling. See https://github.com/Coyote-A/ultimate-upscale-for-automatic1111/issues/111#issuecomment-1663504109 Thanks to Алексей Трофимов and eunnone for reporting the issue. --- modules/processing.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 44d20fb7..63cd025c 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -1275,13 +1275,13 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if self.mask_blur_x > 0: np_mask = np.array(image_mask) - kernel_size = 2 * int(4 * self.mask_blur_x + 0.5) + 1 + kernel_size = 2 * int(2.5 * self.mask_blur_x + 0.5) + 1 np_mask = cv2.GaussianBlur(np_mask, (kernel_size, 1), self.mask_blur_x) image_mask = Image.fromarray(np_mask) if self.mask_blur_y > 0: np_mask = np.array(image_mask) - kernel_size = 2 * int(4 * self.mask_blur_y + 0.5) + 1 + kernel_size = 2 * int(2.5 * self.mask_blur_y + 0.5) + 1 np_mask = cv2.GaussianBlur(np_mask, (1, kernel_size), self.mask_blur_y) image_mask = Image.fromarray(np_mask) -- cgit v1.2.3 From b85ec2b9b66492ff9bf3d40a4d9b424390067f0f Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 5 Aug 2023 13:14:00 +0800 Subject: Fix some merge mistakes --- modules/processing.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b9900ded..43cb763f 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1136,7 +1136,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): batch_images.append(image) decoded_samples = torch.from_numpy(np.array(batch_images)) - decoded_samples = decoded_samples.to(shared.device) + decoded_samples = decoded_samples.to(shared.device, dtype=devices.dtype_vae) self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method samples = images_tensor_to_samples(decoded_samples, approximation_indexes.get(opts.sd_vae_encode_method)) @@ -1374,6 +1374,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): raise RuntimeError(f"bad number of images passed: {len(imgs)}; expecting {self.batch_size} or less") image = torch.from_numpy(batch_images) + image = image.to(shared.device, dtype=devices.dtype_vae) self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() -- cgit v1.2.3 From 3f451f3042cc69a751cebc329c11f9f145da5186 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 5 Aug 2023 10:36:26 +0300 Subject: do not add VAE Encoder/Decoder to infotext if it's the default --- modules/processing.py | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 43cb763f..a9d66005 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -795,7 +795,9 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if getattr(samples_ddim, 'already_decoded', False): x_samples_ddim = samples_ddim else: - p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method + if opts.sd_vae_decode_method != 'Full': + p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method + x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) x_samples_ddim = torch.stack(x_samples_ddim).float() @@ -1138,7 +1140,8 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = torch.from_numpy(np.array(batch_images)) decoded_samples = decoded_samples.to(shared.device, dtype=devices.dtype_vae) - self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method + if opts.sd_vae_encode_method != 'Full': + self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method samples = images_tensor_to_samples(decoded_samples, approximation_indexes.get(opts.sd_vae_encode_method)) image_conditioning = self.img2img_image_conditioning(decoded_samples, samples) @@ -1375,7 +1378,10 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image = torch.from_numpy(batch_images) image = image.to(shared.device, dtype=devices.dtype_vae) - self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method + + if opts.sd_vae_encode_method != 'Full': + self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method + self.init_latent = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) devices.torch_gc() -- cgit v1.2.3 From 1d7dcdb6c38c7bca945b3fa8a5d4a1f93446f22a Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sat, 5 Aug 2023 19:07:39 +0900 Subject: Option to not save incomplete images --- modules/processing.py | 19 +++++++++++-------- modules/shared.py | 1 + 2 files changed, 12 insertions(+), 8 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 43cb763f..bf4f938b 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -103,6 +103,10 @@ def txt2img_image_conditioning(sd_model, x, width, height): return x.new_zeros(x.shape[0], 5, 1, 1, dtype=x.dtype, device=x.device) +def save_images_if_interrupted(): + return not (opts.dont_save_interrupted_images and (state.interrupted or state.skipped)) + + class StableDiffusionProcessing: """ The first set of paramaters: sd_models -> do_not_reload_embeddings represent the minimum required to create a StableDiffusionProcessing @@ -821,6 +825,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: def infotext(index=0, use_main_prompt=False): return create_infotext(p, p.prompts, p.seeds, p.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=p.negative_prompts) + save_images_if_interrupt = save_images_if_interrupted() + for i, x_sample in enumerate(x_samples_ddim): p.batch_index = i @@ -828,7 +834,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: x_sample = x_sample.astype(np.uint8) if p.restore_faces: - if opts.save and not p.do_not_save_samples and opts.save_images_before_face_restoration: + if opts.save and not p.do_not_save_samples and opts.save_images_before_face_restoration and save_images_if_interrupt: images.save_image(Image.fromarray(x_sample), p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-face-restoration") devices.torch_gc() @@ -842,16 +848,15 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: pp = scripts.PostprocessImageArgs(image) p.scripts.postprocess_image(p, pp) image = pp.image - if p.color_corrections is not None and i < len(p.color_corrections): - if opts.save and not p.do_not_save_samples and opts.save_images_before_color_correction: + if opts.save and not p.do_not_save_samples and opts.save_images_before_color_correction and save_images_if_interrupt: image_without_cc = apply_overlay(image, p.paste_to, i, p.overlay_images) images.save_image(image_without_cc, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-color-correction") image = apply_color_correction(p.color_corrections[i], image) image = apply_overlay(image, p.paste_to, i, p.overlay_images) - if opts.samples_save and not p.do_not_save_samples: + if opts.samples_save and not p.do_not_save_samples and save_images_if_interrupt: images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p) text = infotext(i) @@ -859,8 +864,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if opts.enable_pnginfo: image.info["parameters"] = text output_images.append(image) - - if hasattr(p, 'mask_for_overlay') and p.mask_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): + if hasattr(p, 'mask_for_overlay') and p.mask_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]) and save_images_if_interrupt: image_mask = p.mask_for_overlay.convert('RGB') image_mask_composite = Image.composite(image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') @@ -896,7 +900,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: grid.info["parameters"] = text output_images.insert(0, grid) index_of_first_image = 1 - if opts.grid_save: images.save_image(grid, p.outpath_grids, "grid", p.all_seeds[0], p.all_prompts[0], opts.grid_format, info=infotext(use_main_prompt=True), short_filename=not opts.grid_extended_filename, p=p, grid=True) @@ -1091,7 +1094,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def save_intermediate(image, index): """saves image before applying hires fix, if enabled in options; takes as an argument either an image or batch with latent space images""" - if not opts.save or self.do_not_save_samples or not opts.save_images_before_highres_fix: + if not opts.save or self.do_not_save_samples or not opts.save_images_before_highres_fix or not save_images_if_interrupted(): return if not isinstance(image, Image.Image): diff --git a/modules/shared.py b/modules/shared.py index 516ad7e8..a7de686c 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -356,6 +356,7 @@ options_templates.update(options_section(('saving-images', "Saving images/grids" "temp_dir": OptionInfo("", "Directory for temporary images; leave empty for default"), "clean_temp_dir_at_start": OptionInfo(False, "Cleanup non-default temporary directory when starting webui"), + "dont_save_interrupted_images": OptionInfo(False, "Don't save incomplete images").info("Don't save images that has been interrupted in mid-generation, they will still show up in webui output."), })) options_templates.update(options_section(('saving-paths', "Paths for saving"), { -- cgit v1.2.3 From 496cef956b3337c4745b1224e244706bd747195a Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Sat, 5 Aug 2023 21:14:13 -0400 Subject: Allow s_noise override to actually be used --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index a9d66005..1601a14c 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -149,7 +149,7 @@ class StableDiffusionProcessing: self.s_churn = s_churn or opts.s_churn self.s_tmin = s_tmin or opts.s_tmin self.s_tmax = s_tmax or float('inf') # not representable as a standard ui option - self.s_noise = s_noise or opts.s_noise + self.s_noise = opts.data.get('s_noise', s_noise) self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} self.override_settings_restore_afterwards = override_settings_restore_afterwards self.is_using_inpainting_conditioning = False -- cgit v1.2.3 From c11104fed5ffee7b9a22674889580028296c5e55 Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Sat, 5 Aug 2023 21:42:03 -0400 Subject: Add s_tmax --- modules/processing.py | 2 +- modules/shared.py | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index a9d66005..a5cd2a47 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -148,7 +148,7 @@ class StableDiffusionProcessing: self.s_min_uncond = s_min_uncond or opts.s_min_uncond self.s_churn = s_churn or opts.s_churn self.s_tmin = s_tmin or opts.s_tmin - self.s_tmax = s_tmax or float('inf') # not representable as a standard ui option + self.s_tmax = opts.data.get('s_tmax', 0) or float('inf') # not representable as a standard ui option self.s_noise = s_noise or opts.s_noise self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} self.override_settings_restore_afterwards = override_settings_restore_afterwards diff --git a/modules/shared.py b/modules/shared.py index a99b500b..b1c0c0e9 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -608,6 +608,7 @@ options_templates.update(options_section(('sampler-params', "Sampler parameters" "ddim_discretize": OptionInfo('uniform', "img2img DDIM discretize", gr.Radio, {"choices": ['uniform', 'quad']}), 's_churn': OptionInfo(0.0, "sigma churn", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}), 's_tmin': OptionInfo(0.0, "sigma tmin", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}), + 's_tmax': OptionInfo(0.0, "sigma tmax", gr.Slider, {"minimum": 0.0, "maximum": 999.0, "step": 0.01}).info("0 = inf"), 's_noise': OptionInfo(1.0, "sigma noise", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}), 'k_sched_type': OptionInfo("Automatic", "scheduler type", gr.Dropdown, {"choices": ["Automatic", "karras", "exponential", "polyexponential"]}).info("lets you override the noise schedule for k-diffusion samplers; choosing Automatic disables the three parameters below"), 'sigma_min': OptionInfo(0.0, "sigma min", gr.Number).info("0 = default (~0.03); minimum noise strength for k-diffusion noise scheduler"), -- cgit v1.2.3 From 85c2c138d2f5f853c79bde3775d69628e86c65ca Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Sat, 5 Aug 2023 21:51:46 -0400 Subject: Attempt to read s_tmax from arg first if option not found --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index a5cd2a47..2003532d 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -148,7 +148,7 @@ class StableDiffusionProcessing: self.s_min_uncond = s_min_uncond or opts.s_min_uncond self.s_churn = s_churn or opts.s_churn self.s_tmin = s_tmin or opts.s_tmin - self.s_tmax = opts.data.get('s_tmax', 0) or float('inf') # not representable as a standard ui option + self.s_tmax = opts.data.get('s_tmax', s_tmax or 0) or float('inf') # not representable as a standard ui option self.s_noise = s_noise or opts.s_noise self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} self.override_settings_restore_afterwards = override_settings_restore_afterwards -- cgit v1.2.3 From f18a03219039ebaa48ce96adc5b50e0e5ed5d84f Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Sat, 5 Aug 2023 23:05:25 -0400 Subject: Correct s_noise fix --- modules/processing.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 1601a14c..bb268db7 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -110,7 +110,7 @@ class StableDiffusionProcessing: cached_uc = [None, None] cached_c = [None, None] - def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = False, tiling: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = 1.0, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): + def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = False, tiling: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = None, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): if sampler_index is not None: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) @@ -149,7 +149,7 @@ class StableDiffusionProcessing: self.s_churn = s_churn or opts.s_churn self.s_tmin = s_tmin or opts.s_tmin self.s_tmax = s_tmax or float('inf') # not representable as a standard ui option - self.s_noise = opts.data.get('s_noise', s_noise) + self.s_noise = s_noise if s_noise is not None else opts.data.get('s_noise', 1.0) self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} self.override_settings_restore_afterwards = override_settings_restore_afterwards self.is_using_inpainting_conditioning = False -- cgit v1.2.3 From d86d12e9117772f041682124badc7baac7c57911 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 6 Aug 2023 06:21:36 +0300 Subject: rework saving incomplete images --- modules/processing.py | 18 +++++++++--------- modules/shared.py | 2 +- 2 files changed, 10 insertions(+), 10 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 8f26621b..aef8fafd 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -103,10 +103,6 @@ def txt2img_image_conditioning(sd_model, x, width, height): return x.new_zeros(x.shape[0], 5, 1, 1, dtype=x.dtype, device=x.device) -def save_images_if_interrupted(): - return not (opts.dont_save_interrupted_images and (state.interrupted or state.skipped)) - - class StableDiffusionProcessing: """ The first set of paramaters: sd_models -> do_not_reload_embeddings represent the minimum required to create a StableDiffusionProcessing @@ -372,6 +368,10 @@ class StableDiffusionProcessing: def parse_extra_network_prompts(self): self.prompts, self.extra_network_data = extra_networks.parse_prompts(self.prompts) + def save_samples(self) -> bool: + """Returns whether generated images need to be written to disk""" + return opts.samples_save and not self.do_not_save_samples and (opts.save_incomplete_images or not state.interrupted and not state.skipped) + class Processed: def __init__(self, p: StableDiffusionProcessing, images_list, seed=-1, info="", subseed=None, all_prompts=None, all_negative_prompts=None, all_seeds=None, all_subseeds=None, index_of_first_image=0, infotexts=None, comments=""): @@ -827,7 +827,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: def infotext(index=0, use_main_prompt=False): return create_infotext(p, p.prompts, p.seeds, p.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=p.negative_prompts) - save_images_if_interrupt = save_images_if_interrupted() + save_samples = p.save_samples() for i, x_sample in enumerate(x_samples_ddim): p.batch_index = i @@ -836,7 +836,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: x_sample = x_sample.astype(np.uint8) if p.restore_faces: - if opts.save and not p.do_not_save_samples and opts.save_images_before_face_restoration and save_images_if_interrupt: + if save_samples and opts.save_images_before_face_restoration: images.save_image(Image.fromarray(x_sample), p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-face-restoration") devices.torch_gc() @@ -851,14 +851,14 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.scripts.postprocess_image(p, pp) image = pp.image if p.color_corrections is not None and i < len(p.color_corrections): - if opts.save and not p.do_not_save_samples and opts.save_images_before_color_correction and save_images_if_interrupt: + if save_samples and opts.save_images_before_color_correction: image_without_cc = apply_overlay(image, p.paste_to, i, p.overlay_images) images.save_image(image_without_cc, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-color-correction") image = apply_color_correction(p.color_corrections[i], image) image = apply_overlay(image, p.paste_to, i, p.overlay_images) - if opts.samples_save and not p.do_not_save_samples and save_images_if_interrupt: + if save_samples: images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p) text = infotext(i) @@ -1096,7 +1096,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def save_intermediate(image, index): """saves image before applying hires fix, if enabled in options; takes as an argument either an image or batch with latent space images""" - if not opts.save or self.do_not_save_samples or not opts.save_images_before_highres_fix or not save_images_if_interrupted(): + if not self.save_samples() or not opts.save_images_before_highres_fix: return if not isinstance(image, Image.Image): diff --git a/modules/shared.py b/modules/shared.py index 2bd49ff1..3276d45e 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -385,7 +385,7 @@ options_templates.update(options_section(('saving-images', "Saving images/grids" "temp_dir": OptionInfo("", "Directory for temporary images; leave empty for default"), "clean_temp_dir_at_start": OptionInfo(False, "Cleanup non-default temporary directory when starting webui"), - "dont_save_interrupted_images": OptionInfo(False, "Don't save incomplete images").info("Don't save images that has been interrupted in mid-generation, they will still show up in webui output."), + "save_incomplete_images": OptionInfo(False, "Save incomplete images").info("save images that has been interrupted in mid-generation; even if not saved, they will still show up in webui output."), })) options_templates.update(options_section(('saving-paths', "Paths for saving"), { -- cgit v1.2.3 From ee96a6a58852e6db30afac129570efb88e991957 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 6 Aug 2023 06:32:41 +0300 Subject: do the same for s_tmax #12345 --- modules/processing.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 8743ac0a..6625f2de 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -148,8 +148,8 @@ class StableDiffusionProcessing: self.s_min_uncond = s_min_uncond or opts.s_min_uncond self.s_churn = s_churn or opts.s_churn self.s_tmin = s_tmin or opts.s_tmin - self.s_tmax = opts.data.get('s_tmax', s_tmax or 0) or float('inf') # not representable as a standard ui option - self.s_noise = s_noise if s_noise is not None else opts.data.get('s_noise', 1.0) + self.s_tmax = (s_tmax if s_tmax is not None else opts.s_tmax) or float('inf') + self.s_noise = s_noise if s_noise is not None else opts.s_noise self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} self.override_settings_restore_afterwards = override_settings_restore_afterwards self.is_using_inpainting_conditioning = False -- cgit v1.2.3 From 5cae08f2c32b63989aa04fba32e5c56d9ff4fa6e Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 6 Aug 2023 06:55:19 +0300 Subject: fix rework saving incomplete images --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 6625f2de..7d21fb12 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -866,7 +866,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if opts.enable_pnginfo: image.info["parameters"] = text output_images.append(image) - if hasattr(p, 'mask_for_overlay') and p.mask_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]) and save_images_if_interrupt: + if save_samples and hasattr(p, 'mask_for_overlay') and p.mask_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): image_mask = p.mask_for_overlay.convert('RGB') image_mask_composite = Image.composite(image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') -- cgit v1.2.3 From 57e8a11d17a6646fdf551320f5f714fba752987a Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 6 Aug 2023 13:25:51 +0300 Subject: enable cond cache by default --- modules/processing.py | 31 ++++++++++++++++++------------- modules/shared.py | 2 +- 2 files changed, 19 insertions(+), 14 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 7d21fb12..31745006 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -295,7 +295,7 @@ class StableDiffusionProcessing: self.sampler = None self.c = None self.uc = None - if not opts.experimental_persistent_cond_cache: + if not opts.persistent_cond_cache: StableDiffusionProcessing.cached_c = [None, None] StableDiffusionProcessing.cached_uc = [None, None] @@ -319,6 +319,21 @@ class StableDiffusionProcessing: self.all_prompts = [shared.prompt_styles.apply_styles_to_prompt(x, self.styles) for x in self.all_prompts] self.all_negative_prompts = [shared.prompt_styles.apply_negative_styles_to_prompt(x, self.styles) for x in self.all_negative_prompts] + def cached_params(self, required_prompts, steps, extra_network_data): + """Returns parameters that invalidate the cond cache if changed""" + + return ( + required_prompts, + steps, + opts.CLIP_stop_at_last_layers, + shared.sd_model.sd_checkpoint_info, + extra_network_data, + opts.sdxl_crop_left, + opts.sdxl_crop_top, + self.width, + self.height, + ) + def get_conds_with_caching(self, function, required_prompts, steps, caches, extra_network_data): """ Returns the result of calling function(shared.sd_model, required_prompts, steps) @@ -332,17 +347,7 @@ class StableDiffusionProcessing: caches is a list with items described above. """ - cached_params = ( - required_prompts, - steps, - opts.CLIP_stop_at_last_layers, - shared.sd_model.sd_checkpoint_info, - extra_network_data, - opts.sdxl_crop_left, - opts.sdxl_crop_top, - self.width, - self.height, - ) + cached_params = self.cached_params(required_prompts, steps, extra_network_data) for cache in caches: if cache[0] is not None and cached_params == cache[0]: @@ -1184,7 +1189,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): super().close() self.hr_c = None self.hr_uc = None - if not opts.experimental_persistent_cond_cache: + if not opts.persistent_cond_cache: StableDiffusionProcessingTxt2Img.cached_hr_uc = [None, None] StableDiffusionProcessingTxt2Img.cached_hr_c = [None, None] diff --git a/modules/shared.py b/modules/shared.py index 8e1b8063..078e8135 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -506,7 +506,7 @@ options_templates.update(options_section(('optimizations', "Optimizations"), { "token_merging_ratio_img2img": OptionInfo(0.0, "Token merging ratio for img2img", gr.Slider, {"minimum": 0.0, "maximum": 0.9, "step": 0.1}).info("only applies if non-zero and overrides above"), "token_merging_ratio_hr": OptionInfo(0.0, "Token merging ratio for high-res pass", gr.Slider, {"minimum": 0.0, "maximum": 0.9, "step": 0.1}).info("only applies if non-zero and overrides above"), "pad_cond_uncond": OptionInfo(False, "Pad prompt/negative prompt to be same length").info("improves performance when prompt and negative prompt have different lengths; changes seeds"), - "experimental_persistent_cond_cache": OptionInfo(False, "persistent cond cache").info("Experimental, keep cond caches across jobs, reduce overhead."), + "persistent_cond_cache": OptionInfo(True, "Persistent cond cache").info("Do not recalculate conds from prompts if prompts have not changed since previous calculation"), })) options_templates.update(options_section(('compatibility', "Compatibility"), { -- cgit v1.2.3 From f1975b0213f5be400889ec04b3891d1cb571fe20 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 6 Aug 2023 17:01:07 +0300 Subject: initial refiner support --- modules/processing.py | 4 ++++ modules/sd_models.py | 18 +++++++++++++++++- modules/sd_samplers_common.py | 19 ++++++++++++++++++- modules/sd_samplers_compvis.py | 12 +++++++++++- modules/sd_samplers_kdiffusion.py | 30 ++++++++++++++++++++++++------ modules/shared.py | 2 ++ 6 files changed, 76 insertions(+), 9 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 31745006..f4748d6d 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -666,6 +666,10 @@ def process_images(p: StableDiffusionProcessing) -> Processed: stored_opts = {k: opts.data[k] for k in p.override_settings.keys()} try: + # after running refiner, the refiner model is not unloaded - webui swaps back to main model here + if shared.sd_model.sd_checkpoint_info.title != opts.sd_model_checkpoint: + sd_models.reload_model_weights() + # if no checkpoint override or the override checkpoint can't be found, remove override entry and load opts checkpoint if sd_models.checkpoint_aliases.get(p.override_settings.get('sd_model_checkpoint')) is None: p.override_settings.pop('sd_model_checkpoint', None) diff --git a/modules/sd_models.py b/modules/sd_models.py index f6051604..981aa93d 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -289,11 +289,27 @@ def get_checkpoint_state_dict(checkpoint_info: CheckpointInfo, timer): return res +class SkipWritingToConfig: + """This context manager prevents load_model_weights from writing checkpoint name to the config when it loads weight.""" + + skip = False + previous = None + + def __enter__(self): + self.previous = SkipWritingToConfig.skip + SkipWritingToConfig.skip = True + return self + + def __exit__(self, exc_type, exc_value, exc_traceback): + SkipWritingToConfig.skip = self.previous + + def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer): sd_model_hash = checkpoint_info.calculate_shorthash() timer.record("calculate hash") - shared.opts.data["sd_model_checkpoint"] = checkpoint_info.title + if not SkipWritingToConfig.skip: + shared.opts.data["sd_model_checkpoint"] = checkpoint_info.title if state_dict is None: state_dict = get_checkpoint_state_dict(checkpoint_info, timer) diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 39586b40..3f3e83e3 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -2,7 +2,7 @@ from collections import namedtuple import numpy as np import torch from PIL import Image -from modules import devices, images, sd_vae_approx, sd_samplers, sd_vae_taesd, shared +from modules import devices, images, sd_vae_approx, sd_samplers, sd_vae_taesd, shared, sd_models from modules.shared import opts, state SamplerData = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options']) @@ -127,3 +127,20 @@ def replace_torchsde_browinan(): replace_torchsde_browinan() + + +def apply_refiner(sampler): + completed_ratio = sampler.step / sampler.steps + if completed_ratio > shared.opts.sd_refiner_switch_at and shared.sd_model.sd_checkpoint_info.title != shared.opts.sd_refiner_checkpoint: + refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(shared.opts.sd_refiner_checkpoint) + if refiner_checkpoint_info is None: + raise Exception(f'Could not find checkpoint with name {shared.opts.sd_refiner_checkpoint}') + + with sd_models.SkipWritingToConfig(): + sd_models.reload_model_weights(info=refiner_checkpoint_info) + + devices.torch_gc() + + sampler.update_inner_model() + + sampler.p.setup_conds() diff --git a/modules/sd_samplers_compvis.py b/modules/sd_samplers_compvis.py index 4a8396f9..5df926d3 100644 --- a/modules/sd_samplers_compvis.py +++ b/modules/sd_samplers_compvis.py @@ -19,7 +19,8 @@ samplers_data_compvis = [ class VanillaStableDiffusionSampler: def __init__(self, constructor, sd_model): - self.sampler = constructor(sd_model) + self.p = None + self.sampler = constructor(shared.sd_model) self.is_ddim = hasattr(self.sampler, 'p_sample_ddim') self.is_plms = hasattr(self.sampler, 'p_sample_plms') self.is_unipc = isinstance(self.sampler, modules.models.diffusion.uni_pc.UniPCSampler) @@ -32,6 +33,7 @@ class VanillaStableDiffusionSampler: self.nmask = None self.init_latent = None self.sampler_noises = None + self.steps = None self.step = 0 self.stop_at = None self.eta = None @@ -44,6 +46,7 @@ class VanillaStableDiffusionSampler: return 0 def launch_sampling(self, steps, func): + self.steps = steps state.sampling_steps = steps state.sampling_step = 0 @@ -61,10 +64,15 @@ class VanillaStableDiffusionSampler: return res + def update_inner_model(self): + self.sampler.model = shared.sd_model + def before_sample(self, x, ts, cond, unconditional_conditioning): if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException + sd_samplers_common.apply_refiner(self) + if self.stop_at is not None and self.step > self.stop_at: raise sd_samplers_common.InterruptedException @@ -134,6 +142,8 @@ class VanillaStableDiffusionSampler: self.update_step(x) def initialize(self, p): + self.p = p + if self.is_ddim: self.eta = p.eta if p.eta is not None else shared.opts.eta_ddim else: diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index db71a549..be1bd35e 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -2,7 +2,7 @@ from collections import deque import torch import inspect import k_diffusion.sampling -from modules import prompt_parser, devices, sd_samplers_common, sd_samplers_extra +from modules import prompt_parser, devices, sd_samplers_common, sd_samplers_extra, sd_models from modules.processing import StableDiffusionProcessing from modules.shared import opts, state @@ -87,15 +87,25 @@ class CFGDenoiser(torch.nn.Module): negative prompt. """ - def __init__(self, model): + def __init__(self): super().__init__() - self.inner_model = model + self.model_wrap = None self.mask = None self.nmask = None self.init_latent = None + self.steps = None self.step = 0 self.image_cfg_scale = None self.padded_cond_uncond = False + self.p = None + + @property + def inner_model(self): + if self.model_wrap is None: + denoiser = k_diffusion.external.CompVisVDenoiser if shared.sd_model.parameterization == "v" else k_diffusion.external.CompVisDenoiser + self.model_wrap = denoiser(shared.sd_model, quantize=shared.opts.enable_quantization) + + return self.model_wrap def combine_denoised(self, x_out, conds_list, uncond, cond_scale): denoised_uncond = x_out[-uncond.shape[0]:] @@ -113,10 +123,15 @@ class CFGDenoiser(torch.nn.Module): return denoised + def update_inner_model(self): + self.model_wrap = None + def forward(self, x, sigma, uncond, cond, cond_scale, s_min_uncond, image_cond): if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException + sd_samplers_common.apply_refiner(self) + # at self.image_cfg_scale == 1.0 produced results for edit model are the same as with normal sampling, # so is_edit_model is set to False to support AND composition. is_edit_model = shared.sd_model.cond_stage_key == "edit" and self.image_cfg_scale is not None and self.image_cfg_scale != 1.0 @@ -267,13 +282,13 @@ class TorchHijack: class KDiffusionSampler: def __init__(self, funcname, sd_model): - denoiser = k_diffusion.external.CompVisVDenoiser if sd_model.parameterization == "v" else k_diffusion.external.CompVisDenoiser - self.model_wrap = denoiser(sd_model, quantize=shared.opts.enable_quantization) + self.p = None self.funcname = funcname self.func = funcname if callable(funcname) else getattr(k_diffusion.sampling, self.funcname) self.extra_params = sampler_extra_params.get(funcname, []) - self.model_wrap_cfg = CFGDenoiser(self.model_wrap) + self.model_wrap_cfg = CFGDenoiser() + self.model_wrap = self.model_wrap_cfg.inner_model self.sampler_noises = None self.stop_at = None self.eta = None @@ -305,6 +320,7 @@ class KDiffusionSampler: shared.total_tqdm.update() def launch_sampling(self, steps, func): + self.model_wrap_cfg.steps = steps state.sampling_steps = steps state.sampling_step = 0 @@ -324,6 +340,8 @@ class KDiffusionSampler: return p.steps def initialize(self, p: StableDiffusionProcessing): + self.p = p + self.model_wrap_cfg.p = 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.step = 0 diff --git a/modules/shared.py b/modules/shared.py index 078e8135..ed8395dc 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -461,6 +461,8 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "CLIP_stop_at_last_layers": OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}).link("wiki", "https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#clip-skip").info("ignore last layers of CLIP network; 1 ignores none, 2 ignores one layer"), "upcast_attn": OptionInfo(False, "Upcast cross attention layer to float32"), "randn_source": OptionInfo("GPU", "Random number generator source.", gr.Radio, {"choices": ["GPU", "CPU", "NV"]}).info("changes seeds drastically; use CPU to produce the same picture across different videocard vendors; use NV to produce same picture as on NVidia videocards"), + "sd_refiner_checkpoint": OptionInfo(None, "Refiner checkpoint", gr.Dropdown, lambda: {"choices": list_checkpoint_tiles()}, refresh=refresh_checkpoints).info("switch to another model in the middle of generation"), + "sd_refiner_switch_at": OptionInfo(1.0, "Refiner switch at", gr.Slider, {"minimum": 0.01, "maximum": 1.0, "step": 0.01}).info("fraction of sampling steps when the swtch to refiner model should happen; 1=never, 0.5=switch in the middle of generation"), })) options_templates.update(options_section(('sdxl', "Stable Diffusion XL"), { -- cgit v1.2.3 From 5a0db84b6c7322082c7532df11a29a95a59a612b Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 6 Aug 2023 17:53:33 +0300 Subject: add infotext add proper support for recalculating conds in k-diffusion samplers remove support for compvis samplers --- modules/generation_parameters_copypaste.py | 2 ++ modules/processing.py | 10 ++++++++++ modules/sd_samplers_common.py | 29 ++++++++++++++++++++--------- modules/sd_samplers_compvis.py | 2 -- modules/sd_samplers_kdiffusion.py | 24 ++++++++++++++++-------- 5 files changed, 48 insertions(+), 19 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index e71c9601..6711ca16 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -344,6 +344,8 @@ infotext_to_setting_name_mapping = [ ('Pad conds', 'pad_cond_uncond'), ('VAE Encoder', 'sd_vae_encode_method'), ('VAE Decoder', 'sd_vae_decode_method'), + ('Refiner', 'sd_refiner_checkpoint'), + ('Refiner switch at', 'sd_refiner_switch_at'), ] diff --git a/modules/processing.py b/modules/processing.py index f4748d6d..ec66fd8e 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -370,6 +370,9 @@ class StableDiffusionProcessing: self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, self.steps * self.step_multiplier, [self.cached_uc], self.extra_network_data) self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, self.steps * self.step_multiplier, [self.cached_c], self.extra_network_data) + def get_conds(self): + return self.c, self.uc + def parse_extra_network_prompts(self): self.prompts, self.extra_network_data = extra_networks.parse_prompts(self.prompts) @@ -1251,6 +1254,13 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): with devices.autocast(): extra_networks.activate(self, self.extra_network_data) + def get_conds(self): + if self.is_hr_pass: + return self.hr_c, self.hr_uc + + return super().get_conds() + + def parse_extra_network_prompts(self): res = super().parse_extra_network_prompts() diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 3f3e83e3..92bf0ca1 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -131,16 +131,27 @@ replace_torchsde_browinan() def apply_refiner(sampler): completed_ratio = sampler.step / sampler.steps - if completed_ratio > shared.opts.sd_refiner_switch_at and shared.sd_model.sd_checkpoint_info.title != shared.opts.sd_refiner_checkpoint: - refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(shared.opts.sd_refiner_checkpoint) - if refiner_checkpoint_info is None: - raise Exception(f'Could not find checkpoint with name {shared.opts.sd_refiner_checkpoint}') - with sd_models.SkipWritingToConfig(): - sd_models.reload_model_weights(info=refiner_checkpoint_info) + if completed_ratio <= shared.opts.sd_refiner_switch_at: + return False + + if shared.sd_model.sd_checkpoint_info.title == shared.opts.sd_refiner_checkpoint: + return False + + refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(shared.opts.sd_refiner_checkpoint) + if refiner_checkpoint_info is None: + raise Exception(f'Could not find checkpoint with name {shared.opts.sd_refiner_checkpoint}') + + sampler.p.extra_generation_params['Refiner'] = refiner_checkpoint_info.short_title + sampler.p.extra_generation_params['Refiner switch at'] = shared.opts.sd_refiner_switch_at + + with sd_models.SkipWritingToConfig(): + sd_models.reload_model_weights(info=refiner_checkpoint_info) + + devices.torch_gc() + sampler.p.setup_conds() + sampler.update_inner_model() - devices.torch_gc() + return True - sampler.update_inner_model() - sampler.p.setup_conds() diff --git a/modules/sd_samplers_compvis.py b/modules/sd_samplers_compvis.py index 5df926d3..2eeec18a 100644 --- a/modules/sd_samplers_compvis.py +++ b/modules/sd_samplers_compvis.py @@ -71,8 +71,6 @@ class VanillaStableDiffusionSampler: if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException - sd_samplers_common.apply_refiner(self) - if self.stop_at is not None and self.step > self.stop_at: raise sd_samplers_common.InterruptedException diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index 3aee4e3a..46da0a97 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -87,8 +87,9 @@ class CFGDenoiser(torch.nn.Module): negative prompt. """ - def __init__(self): + def __init__(self, sampler): super().__init__() + self.sampler = sampler self.model_wrap = None self.mask = None self.nmask = None @@ -126,11 +127,17 @@ class CFGDenoiser(torch.nn.Module): def update_inner_model(self): self.model_wrap = None + c, uc = self.p.get_conds() + self.sampler.sampler_extra_args['cond'] = c + self.sampler.sampler_extra_args['uncond'] = uc + def forward(self, x, sigma, uncond, cond, cond_scale, s_min_uncond, image_cond): if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException - sd_samplers_common.apply_refiner(self) + if sd_samplers_common.apply_refiner(self): + cond = self.sampler.sampler_extra_args['cond'] + uncond = self.sampler.sampler_extra_args['uncond'] # at self.image_cfg_scale == 1.0 produced results for edit model are the same as with normal sampling, # so is_edit_model is set to False to support AND composition. @@ -282,12 +289,12 @@ class TorchHijack: class KDiffusionSampler: def __init__(self, funcname, sd_model): - self.p = None self.funcname = funcname self.func = funcname if callable(funcname) else getattr(k_diffusion.sampling, self.funcname) self.extra_params = sampler_extra_params.get(funcname, []) - self.model_wrap_cfg = CFGDenoiser() + self.sampler_extra_args = {} + self.model_wrap_cfg = CFGDenoiser(self) self.model_wrap = self.model_wrap_cfg.inner_model self.sampler_noises = None self.stop_at = None @@ -476,7 +483,7 @@ class KDiffusionSampler: self.model_wrap_cfg.init_latent = x self.last_latent = x - extra_args = { + self.sampler_extra_args = { 'cond': conditioning, 'image_cond': image_conditioning, 'uncond': unconditional_conditioning, @@ -484,7 +491,7 @@ class KDiffusionSampler: 's_min_uncond': self.s_min_uncond } - samples = self.launch_sampling(t_enc + 1, lambda: self.func(self.model_wrap_cfg, xi, extra_args=extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs)) + samples = self.launch_sampling(t_enc + 1, lambda: self.func(self.model_wrap_cfg, xi, extra_args=self.sampler_extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs)) if self.model_wrap_cfg.padded_cond_uncond: p.extra_generation_params["Pad conds"] = True @@ -514,13 +521,14 @@ class KDiffusionSampler: extra_params_kwargs['noise_sampler'] = noise_sampler self.last_latent = x - samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args={ + self.sampler_extra_args = { 'cond': conditioning, 'image_cond': image_conditioning, 'uncond': unconditional_conditioning, 'cond_scale': p.cfg_scale, 's_min_uncond': self.s_min_uncond - }, disable=False, callback=self.callback_state, **extra_params_kwargs)) + } + samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args=self.sampler_extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs)) if self.model_wrap_cfg.padded_cond_uncond: p.extra_generation_params["Pad conds"] = True -- cgit v1.2.3 From 2a72d76d6f3d34b1ffccec7736b19e7d52033dad Mon Sep 17 00:00:00 2001 From: dhwz Date: Tue, 8 Aug 2023 19:08:37 +0200 Subject: fix typo --- modules/processing.py | 2 +- modules/shared.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 31745006..dc6e8ff1 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -568,7 +568,7 @@ def decode_latent_batch(model, batch, target_device=None, check_for_nans=False): errors.print_error_explanation( "A tensor with all NaNs was produced in VAE.\n" "Web UI will now convert VAE into 32-bit float and retry.\n" - "To disable this behavior, disable the 'Automaticlly revert VAE to 32-bit floats' setting.\n" + "To disable this behavior, disable the 'Automatically revert VAE to 32-bit floats' setting.\n" "To always start with 32-bit VAE, use --no-half-vae commandline flag." ) diff --git a/modules/shared.py b/modules/shared.py index 97f1eab5..e34847ce 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -481,7 +481,7 @@ For img2img, VAE is used to process user's input image before the sampling, and "sd_vae_checkpoint_cache": OptionInfo(0, "VAE Checkpoints to cache in RAM", gr.Slider, {"minimum": 0, "maximum": 10, "step": 1}), "sd_vae": OptionInfo("Automatic", "SD VAE", gr.Dropdown, lambda: {"choices": shared_items.sd_vae_items()}, refresh=shared_items.refresh_vae_list).info("choose VAE model: Automatic = use one with same filename as checkpoint; None = use VAE from checkpoint"), "sd_vae_overrides_per_model_preferences": OptionInfo(True, "Selected VAE overrides per-model preferences").info("you can set per-model VAE either by editing user metadata for checkpoints, or by making the VAE have same name as checkpoint"), - "auto_vae_precision": OptionInfo(True, "Automaticlly revert VAE to 32-bit floats").info("triggers when a tensor with NaNs is produced in VAE; disabling the option in this case will result in a black square image"), + "auto_vae_precision": OptionInfo(True, "Automatically revert VAE to 32-bit floats").info("triggers when a tensor with NaNs is produced in VAE; disabling the option in this case will result in a black square image"), "sd_vae_encode_method": OptionInfo("Full", "VAE type for encode", gr.Radio, {"choices": ["Full", "TAESD"]}).info("method to encode image to latent (use in img2img, hires-fix or inpaint mask)"), "sd_vae_decode_method": OptionInfo("Full", "VAE type for decode", gr.Radio, {"choices": ["Full", "TAESD"]}).info("method to decode latent to image"), })) -- cgit v1.2.3 From a8a256f9b5b445206818bfc8a363ed5a1ba50c86 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 8 Aug 2023 21:07:18 +0300 Subject: REMOVE --- modules/processing.py | 3 - modules/sd_hijack.py | 4 +- modules/sd_hijack_inpainting.py | 95 --------------- modules/sd_samplers.py | 7 +- modules/sd_samplers_cfg_denoiser.py | 1 - modules/sd_samplers_compvis.py | 224 ------------------------------------ modules/sd_samplers_kdiffusion.py | 3 +- modules/sd_samplers_timesteps.py | 6 +- 8 files changed, 7 insertions(+), 336 deletions(-) delete mode 100644 modules/sd_hijack_inpainting.py delete mode 100644 modules/sd_samplers_compvis.py (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 31745006..61ba5f11 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1112,9 +1112,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): img2img_sampler_name = self.hr_sampler_name or self.sampler_name - if self.sampler_name in ['PLMS', 'UniPC']: # PLMS/UniPC do not support img2img so we just silently switch to DDIM - img2img_sampler_name = 'DDIM' - self.sampler = sd_samplers.create_sampler(img2img_sampler_name, self.sd_model) if self.latent_scale_mode is not None: diff --git a/modules/sd_hijack.py b/modules/sd_hijack.py index 9ad98199..46652fbd 100644 --- a/modules/sd_hijack.py +++ b/modules/sd_hijack.py @@ -5,7 +5,7 @@ from types import MethodType from modules import devices, sd_hijack_optimizations, shared, script_callbacks, errors, sd_unet from modules.hypernetworks import hypernetwork from modules.shared import cmd_opts -from modules import sd_hijack_clip, sd_hijack_open_clip, sd_hijack_unet, sd_hijack_xlmr, xlmr, sd_hijack_inpainting +from modules import sd_hijack_clip, sd_hijack_open_clip, sd_hijack_unet, sd_hijack_xlmr, xlmr import ldm.modules.attention import ldm.modules.diffusionmodules.model @@ -34,8 +34,6 @@ ldm.modules.diffusionmodules.model.print = shared.ldm_print ldm.util.print = shared.ldm_print ldm.models.diffusion.ddpm.print = shared.ldm_print -sd_hijack_inpainting.do_inpainting_hijack() - optimizers = [] current_optimizer: sd_hijack_optimizations.SdOptimization = None diff --git a/modules/sd_hijack_inpainting.py b/modules/sd_hijack_inpainting.py deleted file mode 100644 index 2d44b856..00000000 --- a/modules/sd_hijack_inpainting.py +++ /dev/null @@ -1,95 +0,0 @@ -import torch - -import ldm.models.diffusion.ddpm -import ldm.models.diffusion.ddim -import ldm.models.diffusion.plms - -from ldm.models.diffusion.ddim import noise_like -from ldm.models.diffusion.sampling_util import norm_thresholding - - -@torch.no_grad() -def p_sample_plms(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False, - temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None, - unconditional_guidance_scale=1., unconditional_conditioning=None, old_eps=None, t_next=None, dynamic_threshold=None): - b, *_, device = *x.shape, x.device - - def get_model_output(x, t): - if unconditional_conditioning is None or unconditional_guidance_scale == 1.: - e_t = self.model.apply_model(x, t, c) - else: - x_in = torch.cat([x] * 2) - t_in = torch.cat([t] * 2) - - if isinstance(c, dict): - assert isinstance(unconditional_conditioning, dict) - c_in = {} - for k in c: - if isinstance(c[k], list): - c_in[k] = [ - torch.cat([unconditional_conditioning[k][i], c[k][i]]) - for i in range(len(c[k])) - ] - else: - c_in[k] = torch.cat([unconditional_conditioning[k], c[k]]) - else: - c_in = torch.cat([unconditional_conditioning, c]) - - e_t_uncond, e_t = self.model.apply_model(x_in, t_in, c_in).chunk(2) - e_t = e_t_uncond + unconditional_guidance_scale * (e_t - e_t_uncond) - - if score_corrector is not None: - assert self.model.parameterization == "eps" - e_t = score_corrector.modify_score(self.model, e_t, x, t, c, **corrector_kwargs) - - return e_t - - alphas = self.model.alphas_cumprod if use_original_steps else self.ddim_alphas - alphas_prev = self.model.alphas_cumprod_prev if use_original_steps else self.ddim_alphas_prev - sqrt_one_minus_alphas = self.model.sqrt_one_minus_alphas_cumprod if use_original_steps else self.ddim_sqrt_one_minus_alphas - sigmas = self.model.ddim_sigmas_for_original_num_steps if use_original_steps else self.ddim_sigmas - - def get_x_prev_and_pred_x0(e_t, index): - # select parameters corresponding to the currently considered timestep - a_t = torch.full((b, 1, 1, 1), alphas[index], device=device) - a_prev = torch.full((b, 1, 1, 1), alphas_prev[index], device=device) - sigma_t = torch.full((b, 1, 1, 1), sigmas[index], device=device) - sqrt_one_minus_at = torch.full((b, 1, 1, 1), sqrt_one_minus_alphas[index],device=device) - - # current prediction for x_0 - pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt() - if quantize_denoised: - pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0) - if dynamic_threshold is not None: - pred_x0 = norm_thresholding(pred_x0, dynamic_threshold) - # direction pointing to x_t - dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t - noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature - if noise_dropout > 0.: - noise = torch.nn.functional.dropout(noise, p=noise_dropout) - x_prev = a_prev.sqrt() * pred_x0 + dir_xt + noise - return x_prev, pred_x0 - - e_t = get_model_output(x, t) - if len(old_eps) == 0: - # Pseudo Improved Euler (2nd order) - x_prev, pred_x0 = get_x_prev_and_pred_x0(e_t, index) - e_t_next = get_model_output(x_prev, t_next) - e_t_prime = (e_t + e_t_next) / 2 - elif len(old_eps) == 1: - # 2nd order Pseudo Linear Multistep (Adams-Bashforth) - e_t_prime = (3 * e_t - old_eps[-1]) / 2 - elif len(old_eps) == 2: - # 3nd order Pseudo Linear Multistep (Adams-Bashforth) - e_t_prime = (23 * e_t - 16 * old_eps[-1] + 5 * old_eps[-2]) / 12 - elif len(old_eps) >= 3: - # 4nd order Pseudo Linear Multistep (Adams-Bashforth) - e_t_prime = (55 * e_t - 59 * old_eps[-1] + 37 * old_eps[-2] - 9 * old_eps[-3]) / 24 - - x_prev, pred_x0 = get_x_prev_and_pred_x0(e_t_prime, index) - - return x_prev, pred_x0, e_t - - -def do_inpainting_hijack(): - ldm.models.diffusion.plms.PLMSSampler.p_sample_plms = p_sample_plms diff --git a/modules/sd_samplers.py b/modules/sd_samplers.py index fe206894..05dbe2b5 100644 --- a/modules/sd_samplers.py +++ b/modules/sd_samplers.py @@ -1,11 +1,10 @@ -from modules import sd_samplers_compvis, sd_samplers_kdiffusion, sd_samplers_timesteps, shared +from modules import sd_samplers_kdiffusion, sd_samplers_timesteps, shared # imports for functions that previously were here and are used by other modules from modules.sd_samplers_common import samples_to_image_grid, sample_to_image # noqa: F401 all_samplers = [ *sd_samplers_kdiffusion.samplers_data_k_diffusion, - *sd_samplers_compvis.samplers_data_compvis, *sd_samplers_timesteps.samplers_data_timesteps, ] all_samplers_map = {x.name: x for x in all_samplers} @@ -42,10 +41,8 @@ def set_samplers(): global samplers, samplers_for_img2img hidden = set(shared.opts.hide_samplers) - hidden_img2img = set(shared.opts.hide_samplers + ['PLMS', 'UniPC']) - samplers = [x for x in all_samplers if x.name not in hidden] - samplers_for_img2img = [x for x in all_samplers if x.name not in hidden_img2img] + samplers_for_img2img = [x for x in all_samplers if x.name not in hidden] samplers_map.clear() for sampler in all_samplers: diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index 166a00c7..d826222c 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -1,4 +1,3 @@ -from collections import deque import torch from modules import prompt_parser, devices, sd_samplers_common diff --git a/modules/sd_samplers_compvis.py b/modules/sd_samplers_compvis.py deleted file mode 100644 index 4a8396f9..00000000 --- a/modules/sd_samplers_compvis.py +++ /dev/null @@ -1,224 +0,0 @@ -import math -import ldm.models.diffusion.ddim -import ldm.models.diffusion.plms - -import numpy as np -import torch - -from modules.shared import state -from modules import sd_samplers_common, prompt_parser, shared -import modules.models.diffusion.uni_pc - - -samplers_data_compvis = [ - sd_samplers_common.SamplerData('DDIM', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.ddim.DDIMSampler, model), [], {"default_eta_is_0": True, "uses_ensd": True, "no_sdxl": True}), - sd_samplers_common.SamplerData('PLMS', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.plms.PLMSSampler, model), [], {"no_sdxl": True}), - sd_samplers_common.SamplerData('UniPC', lambda model: VanillaStableDiffusionSampler(modules.models.diffusion.uni_pc.UniPCSampler, model), [], {"no_sdxl": True}), -] - - -class VanillaStableDiffusionSampler: - def __init__(self, constructor, sd_model): - self.sampler = constructor(sd_model) - self.is_ddim = hasattr(self.sampler, 'p_sample_ddim') - self.is_plms = hasattr(self.sampler, 'p_sample_plms') - self.is_unipc = isinstance(self.sampler, modules.models.diffusion.uni_pc.UniPCSampler) - self.orig_p_sample_ddim = None - if self.is_plms: - self.orig_p_sample_ddim = self.sampler.p_sample_plms - elif self.is_ddim: - self.orig_p_sample_ddim = self.sampler.p_sample_ddim - self.mask = None - self.nmask = None - self.init_latent = None - self.sampler_noises = None - self.step = 0 - self.stop_at = None - self.eta = None - self.config = None - self.last_latent = None - - self.conditioning_key = sd_model.model.conditioning_key - - def number_of_needed_noises(self, p): - return 0 - - def launch_sampling(self, steps, func): - state.sampling_steps = steps - state.sampling_step = 0 - - try: - return func() - except sd_samplers_common.InterruptedException: - return self.last_latent - - def p_sample_ddim_hook(self, x_dec, cond, ts, unconditional_conditioning, *args, **kwargs): - x_dec, ts, cond, unconditional_conditioning = self.before_sample(x_dec, ts, cond, unconditional_conditioning) - - res = self.orig_p_sample_ddim(x_dec, cond, ts, *args, unconditional_conditioning=unconditional_conditioning, **kwargs) - - x_dec, ts, cond, unconditional_conditioning, res = self.after_sample(x_dec, ts, cond, unconditional_conditioning, res) - - return res - - def before_sample(self, x, ts, cond, unconditional_conditioning): - if state.interrupted or state.skipped: - raise sd_samplers_common.InterruptedException - - if self.stop_at is not None and self.step > self.stop_at: - raise sd_samplers_common.InterruptedException - - # Have to unwrap the inpainting conditioning here to perform pre-processing - image_conditioning = None - uc_image_conditioning = None - if isinstance(cond, dict): - if self.conditioning_key == "crossattn-adm": - image_conditioning = cond["c_adm"] - uc_image_conditioning = unconditional_conditioning["c_adm"] - else: - image_conditioning = cond["c_concat"][0] - cond = cond["c_crossattn"][0] - unconditional_conditioning = unconditional_conditioning["c_crossattn"][0] - - conds_list, tensor = prompt_parser.reconstruct_multicond_batch(cond, self.step) - unconditional_conditioning = prompt_parser.reconstruct_cond_batch(unconditional_conditioning, self.step) - - assert all(len(conds) == 1 for conds in conds_list), 'composition via AND is not supported for DDIM/PLMS samplers' - cond = tensor - - # for DDIM, shapes must match, we can't just process cond and uncond independently; - # filling unconditional_conditioning with repeats of the last vector to match length is - # not 100% correct but should work well enough - if unconditional_conditioning.shape[1] < cond.shape[1]: - last_vector = unconditional_conditioning[:, -1:] - last_vector_repeated = last_vector.repeat([1, cond.shape[1] - unconditional_conditioning.shape[1], 1]) - unconditional_conditioning = torch.hstack([unconditional_conditioning, last_vector_repeated]) - elif unconditional_conditioning.shape[1] > cond.shape[1]: - unconditional_conditioning = unconditional_conditioning[:, :cond.shape[1]] - - if self.mask is not None: - img_orig = self.sampler.model.q_sample(self.init_latent, ts) - x = img_orig * self.mask + self.nmask * x - - # Wrap the image conditioning back up since the DDIM code can accept the dict directly. - # Note that they need to be lists because it just concatenates them later. - if image_conditioning is not None: - if self.conditioning_key == "crossattn-adm": - cond = {"c_adm": image_conditioning, "c_crossattn": [cond]} - unconditional_conditioning = {"c_adm": uc_image_conditioning, "c_crossattn": [unconditional_conditioning]} - else: - cond = {"c_concat": [image_conditioning], "c_crossattn": [cond]} - unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]} - - return x, ts, cond, unconditional_conditioning - - def update_step(self, last_latent): - if self.mask is not None: - self.last_latent = self.init_latent * self.mask + self.nmask * last_latent - else: - self.last_latent = last_latent - - sd_samplers_common.store_latent(self.last_latent) - - self.step += 1 - state.sampling_step = self.step - shared.total_tqdm.update() - - def after_sample(self, x, ts, cond, uncond, res): - if not self.is_unipc: - self.update_step(res[1]) - - return x, ts, cond, uncond, res - - def unipc_after_update(self, x, model_x): - self.update_step(x) - - def initialize(self, p): - if self.is_ddim: - self.eta = p.eta if p.eta is not None else shared.opts.eta_ddim - else: - self.eta = 0.0 - - if self.eta != 0.0: - p.extra_generation_params["Eta DDIM"] = self.eta - - if self.is_unipc: - keys = [ - ('UniPC variant', 'uni_pc_variant'), - ('UniPC skip type', 'uni_pc_skip_type'), - ('UniPC order', 'uni_pc_order'), - ('UniPC lower order final', 'uni_pc_lower_order_final'), - ] - - for name, key in keys: - v = getattr(shared.opts, key) - if v != shared.opts.get_default(key): - p.extra_generation_params[name] = v - - for fieldname in ['p_sample_ddim', 'p_sample_plms']: - if hasattr(self.sampler, fieldname): - setattr(self.sampler, fieldname, self.p_sample_ddim_hook) - if self.is_unipc: - self.sampler.set_hooks(lambda x, t, c, u: self.before_sample(x, t, c, u), lambda x, t, c, u, r: self.after_sample(x, t, c, u, r), lambda x, mx: self.unipc_after_update(x, mx)) - - self.mask = p.mask if hasattr(p, 'mask') else None - self.nmask = p.nmask if hasattr(p, 'nmask') else None - - - def adjust_steps_if_invalid(self, p, num_steps): - if ((self.config.name == 'DDIM') and p.ddim_discretize == 'uniform') or (self.config.name == 'PLMS') or (self.config.name == 'UniPC'): - if self.config.name == 'UniPC' and num_steps < shared.opts.uni_pc_order: - num_steps = shared.opts.uni_pc_order - valid_step = 999 / (1000 // num_steps) - if valid_step == math.floor(valid_step): - return int(valid_step) + 1 - - return num_steps - - def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None): - steps, t_enc = sd_samplers_common.setup_img2img_steps(p, steps) - steps = self.adjust_steps_if_invalid(p, steps) - self.initialize(p) - - self.sampler.make_schedule(ddim_num_steps=steps, 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.init_latent = x - self.last_latent = x - self.step = 0 - - # Wrap the conditioning models with additional image conditioning for inpainting model - if image_conditioning is not None: - if self.conditioning_key == "crossattn-adm": - conditioning = {"c_adm": image_conditioning, "c_crossattn": [conditioning]} - unconditional_conditioning = {"c_adm": torch.zeros_like(image_conditioning), "c_crossattn": [unconditional_conditioning]} - else: - conditioning = {"c_concat": [image_conditioning], "c_crossattn": [conditioning]} - unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]} - - samples = self.launch_sampling(t_enc + 1, lambda: self.sampler.decode(x1, conditioning, t_enc, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning)) - - return samples - - def sample(self, p, x, conditioning, unconditional_conditioning, steps=None, image_conditioning=None): - self.initialize(p) - - self.init_latent = None - self.last_latent = x - self.step = 0 - - steps = self.adjust_steps_if_invalid(p, steps or p.steps) - - # Wrap the conditioning models with additional image conditioning for inpainting model - # dummy_for_plms is needed because PLMS code checks the first item in the dict to have the right shape - if image_conditioning is not None: - if self.conditioning_key == "crossattn-adm": - conditioning = {"dummy_for_plms": np.zeros((conditioning.shape[0],)), "c_crossattn": [conditioning], "c_adm": image_conditioning} - unconditional_conditioning = {"c_crossattn": [unconditional_conditioning], "c_adm": torch.zeros_like(image_conditioning)} - else: - conditioning = {"dummy_for_plms": np.zeros((conditioning.shape[0],)), "c_crossattn": [conditioning], "c_concat": [image_conditioning]} - unconditional_conditioning = {"c_crossattn": [unconditional_conditioning], "c_concat": [image_conditioning]} - - samples_ddim = self.launch_sampling(steps, lambda: 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)[0]) - - return samples_ddim diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index 3a2e01b7..27a73486 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -1,8 +1,7 @@ -from collections import deque import torch import inspect import k_diffusion.sampling -from modules import devices, sd_samplers_common, sd_samplers_extra, sd_samplers_cfg_denoiser +from modules import sd_samplers_common, sd_samplers_extra, sd_samplers_cfg_denoiser from modules.shared import opts import modules.shared as shared diff --git a/modules/sd_samplers_timesteps.py b/modules/sd_samplers_timesteps.py index 8560d009..d89d0efb 100644 --- a/modules/sd_samplers_timesteps.py +++ b/modules/sd_samplers_timesteps.py @@ -7,9 +7,9 @@ from modules.shared import opts import modules.shared as shared samplers_timesteps = [ - ('k_DDIM', sd_samplers_timesteps_impl.ddim, ['k_ddim'], {}), - ('k_PLMS', sd_samplers_timesteps_impl.plms, ['k_plms'], {}), - ('k_UniPC', sd_samplers_timesteps_impl.unipc, ['k_unipc'], {}), + ('DDIM', sd_samplers_timesteps_impl.ddim, ['ddim'], {}), + ('PLMS', sd_samplers_timesteps_impl.plms, ['plms'], {}), + ('UniPC', sd_samplers_timesteps_impl.unipc, ['unipc'], {}), ] -- cgit v1.2.3 From ec194b637476855ea5918a44a65e85fb587483ab Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 8 Aug 2023 22:14:02 +0300 Subject: fix webui not switching back to original model from refiner when batch count is greater than 1 --- modules/processing.py | 2 ++ 1 file changed, 2 insertions(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b635cc74..cf62cdd3 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -766,6 +766,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if state.interrupted: break + sd_models.reload_model_weights() # model can be changed for example by refiner + p.prompts = p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size] p.negative_prompts = p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size] p.seeds = p.all_seeds[n * p.batch_size:(n + 1) * p.batch_size] -- cgit v1.2.3 From d81d3fa8cde83ce1421889ed481a69c950c0c6f6 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 9 Aug 2023 07:45:06 +0300 Subject: fix styles missing from the prompt in infotext when making a grid of batch of multiplie images --- modules/processing.py | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index d7266307..aa72b132 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -172,6 +172,8 @@ class StableDiffusionProcessing: self.iteration = 0 self.is_hr_pass = False self.sampler = None + self.main_prompt = None + self.main_negative_prompt = None self.prompts = None self.negative_prompts = None @@ -319,6 +321,9 @@ class StableDiffusionProcessing: self.all_prompts = [shared.prompt_styles.apply_styles_to_prompt(x, self.styles) for x in self.all_prompts] self.all_negative_prompts = [shared.prompt_styles.apply_negative_styles_to_prompt(x, self.styles) for x in self.all_negative_prompts] + self.main_prompt = self.all_prompts[0] + self.main_negative_prompt = self.all_negative_prompts[0] + def cached_params(self, required_prompts, steps, extra_network_data): """Returns parameters that invalidate the cond cache if changed""" @@ -653,8 +658,8 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter generation_params_text = ", ".join([k if k == v else f'{k}: {generation_parameters_copypaste.quote(v)}' for k, v in generation_params.items() if v is not None]) - prompt_text = p.prompt if use_main_prompt else all_prompts[index] - negative_prompt_text = f"\nNegative prompt: {all_negative_prompts[index]}" if all_negative_prompts[index] else "" + prompt_text = p.main_prompt if use_main_prompt else all_prompts[index] + negative_prompt_text = f"\nNegative prompt: {p.main_negative_prompt if use_main_prompt else all_negative_prompts[index]}" if all_negative_prompts[index] else "" return f"{prompt_text}{negative_prompt_text}\n{generation_params_text}".strip() -- cgit v1.2.3 From 0d5dc9a6e7f6362e423a06bf0e75dd5854025394 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 9 Aug 2023 08:43:31 +0300 Subject: rework RNG to use generators instead of generating noises beforehand --- modules/devices.py | 81 +------------------- modules/processing.py | 89 +++------------------- modules/rng.py | 171 ++++++++++++++++++++++++++++++++++++++++++ modules/sd_samplers_common.py | 24 +++--- modules/shared.py | 2 +- 5 files changed, 196 insertions(+), 171 deletions(-) create mode 100644 modules/rng.py (limited to 'modules/processing.py') diff --git a/modules/devices.py b/modules/devices.py index 00a00b18..ce59dc53 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -3,7 +3,7 @@ import contextlib from functools import lru_cache import torch -from modules import errors, rng_philox +from modules import errors if sys.platform == "darwin": from modules import mac_specific @@ -96,84 +96,6 @@ def cond_cast_float(input): nv_rng = None -def randn(seed, shape): - """Generate a tensor with random numbers from a normal distribution using seed. - - Uses the seed parameter to set the global torch seed; to generate more with that seed, use randn_like/randn_without_seed.""" - - from modules.shared import opts - - manual_seed(seed) - - if opts.randn_source == "NV": - return torch.asarray(nv_rng.randn(shape), device=device) - - if opts.randn_source == "CPU" or device.type == 'mps': - return torch.randn(shape, device=cpu).to(device) - - return torch.randn(shape, device=device) - - -def randn_local(seed, shape): - """Generate a tensor with random numbers from a normal distribution using seed. - - Does not change the global random number generator. You can only generate the seed's first tensor using this function.""" - - from modules.shared import opts - - if opts.randn_source == "NV": - rng = rng_philox.Generator(seed) - return torch.asarray(rng.randn(shape), device=device) - - local_device = cpu if opts.randn_source == "CPU" or device.type == 'mps' else device - local_generator = torch.Generator(local_device).manual_seed(int(seed)) - return torch.randn(shape, device=local_device, generator=local_generator).to(device) - - -def randn_like(x): - """Generate a tensor with random numbers from a normal distribution using the previously initialized genrator. - - Use either randn() or manual_seed() to initialize the generator.""" - - from modules.shared import opts - - if opts.randn_source == "NV": - return torch.asarray(nv_rng.randn(x.shape), device=x.device, dtype=x.dtype) - - if opts.randn_source == "CPU" or x.device.type == 'mps': - return torch.randn_like(x, device=cpu).to(x.device) - - return torch.randn_like(x) - - -def randn_without_seed(shape): - """Generate a tensor with random numbers from a normal distribution using the previously initialized genrator. - - Use either randn() or manual_seed() to initialize the generator.""" - - from modules.shared import opts - - if opts.randn_source == "NV": - return torch.asarray(nv_rng.randn(shape), device=device) - - if opts.randn_source == "CPU" or device.type == 'mps': - return torch.randn(shape, device=cpu).to(device) - - return torch.randn(shape, device=device) - - -def manual_seed(seed): - """Set up a global random number generator using the specified seed.""" - from modules.shared import opts - - if opts.randn_source == "NV": - global nv_rng - nv_rng = rng_philox.Generator(seed) - return - - torch.manual_seed(seed) - - def autocast(disable=False): from modules import shared @@ -236,3 +158,4 @@ def first_time_calculation(): x = torch.zeros((1, 1, 3, 3)).to(device, dtype) conv2d = torch.nn.Conv2d(1, 1, (3, 3)).to(device, dtype) conv2d(x) + diff --git a/modules/processing.py b/modules/processing.py index aa72b132..2df5e8c7 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -14,7 +14,7 @@ from skimage import exposure from typing import Any, Dict, List import modules.sd_hijack -from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors +from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng from modules.sd_hijack import model_hijack from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes from modules.shared import opts, cmd_opts, state @@ -186,6 +186,7 @@ class StableDiffusionProcessing: self.cached_c = StableDiffusionProcessing.cached_c self.uc = None self.c = None + self.rng: rng.ImageRNG = None self.user = None @@ -475,82 +476,9 @@ class Processed: return self.token_merging_ratio_hr if for_hr else self.token_merging_ratio -# from https://discuss.pytorch.org/t/help-regarding-slerp-function-for-generative-model-sampling/32475/3 -def slerp(val, low, high): - low_norm = low/torch.norm(low, dim=1, keepdim=True) - high_norm = high/torch.norm(high, dim=1, keepdim=True) - dot = (low_norm*high_norm).sum(1) - - if dot.mean() > 0.9995: - return low * val + high * (1 - val) - - omega = torch.acos(dot) - so = torch.sin(omega) - res = (torch.sin((1.0-val)*omega)/so).unsqueeze(1)*low + (torch.sin(val*omega)/so).unsqueeze(1) * high - return res - - def create_random_tensors(shape, seeds, subseeds=None, subseed_strength=0.0, seed_resize_from_h=0, seed_resize_from_w=0, p=None): - eta_noise_seed_delta = opts.eta_noise_seed_delta or 0 - xs = [] - - # if we have multiple seeds, this means we are working with batch size>1; this then - # enables the generation of additional tensors with noise that the sampler will use during its processing. - # Using those pre-generated tensors instead of simple torch.randn allows a batch with seeds [100, 101] to - # produce the same images as with two batches [100], [101]. - if p is not None and p.sampler is not None and (len(seeds) > 1 and opts.enable_batch_seeds or eta_noise_seed_delta > 0): - sampler_noises = [[] for _ in range(p.sampler.number_of_needed_noises(p))] - else: - sampler_noises = None - - for i, seed in enumerate(seeds): - noise_shape = shape if seed_resize_from_h <= 0 or seed_resize_from_w <= 0 else (shape[0], seed_resize_from_h//8, seed_resize_from_w//8) - - subnoise = None - if subseeds is not None and subseed_strength != 0: - subseed = 0 if i >= len(subseeds) else subseeds[i] - - subnoise = devices.randn(subseed, noise_shape) - - # randn results depend on device; gpu and cpu get different results for same seed; - # the way I see it, it's better to do this on CPU, so that everyone gets same result; - # but the original script had it like this, so I do not dare change it for now because - # it will break everyone's seeds. - noise = devices.randn(seed, noise_shape) - - if subnoise is not None: - noise = slerp(subseed_strength, noise, subnoise) - - if noise_shape != shape: - x = devices.randn(seed, shape) - dx = (shape[2] - noise_shape[2]) // 2 - dy = (shape[1] - noise_shape[1]) // 2 - w = noise_shape[2] if dx >= 0 else noise_shape[2] + 2 * dx - h = noise_shape[1] if dy >= 0 else noise_shape[1] + 2 * dy - tx = 0 if dx < 0 else dx - ty = 0 if dy < 0 else dy - dx = max(-dx, 0) - dy = max(-dy, 0) - - x[:, ty:ty+h, tx:tx+w] = noise[:, dy:dy+h, dx:dx+w] - noise = x - - if sampler_noises is not None: - cnt = p.sampler.number_of_needed_noises(p) - - if eta_noise_seed_delta > 0: - devices.manual_seed(seed + eta_noise_seed_delta) - - for j in range(cnt): - sampler_noises[j].append(devices.randn_without_seed(tuple(noise_shape))) - - xs.append(noise) - - if sampler_noises is not None: - p.sampler.sampler_noises = [torch.stack(n).to(shared.device) for n in sampler_noises] - - x = torch.stack(xs).to(shared.device) - return x + g = rng.ImageRNG(shape, seeds, subseeds=subseeds, subseed_strength=subseed_strength, seed_resize_from_h=seed_resize_from_h, seed_resize_from_w=seed_resize_from_w) + return g.next() class DecodedSamples(list): @@ -769,6 +697,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.seeds = p.all_seeds[n * p.batch_size:(n + 1) * p.batch_size] p.subseeds = p.all_subseeds[n * p.batch_size:(n + 1) * p.batch_size] + p.rng = rng.ImageRNG((opt_C, p.height // opt_f, p.width // opt_f), p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, seed_resize_from_h=p.seed_resize_from_h, seed_resize_from_w=p.seed_resize_from_w) + if p.scripts is not None: p.scripts.before_process_batch(p, batch_number=n, prompts=p.prompts, seeds=p.seeds, subseeds=p.subseeds) @@ -1072,7 +1002,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) - x = create_random_tensors([opt_C, self.height // opt_f, self.width // opt_f], seeds=seeds, subseeds=subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self) + x = self.rng.next() samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x)) del x @@ -1160,7 +1090,8 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): samples = samples[:, :, self.truncate_y//2:samples.shape[2]-(self.truncate_y+1)//2, self.truncate_x//2:samples.shape[3]-(self.truncate_x+1)//2] - noise = create_random_tensors(samples.shape[1:], seeds=seeds, subseeds=subseeds, subseed_strength=subseed_strength, p=self) + self.rng = rng.ImageRNG(samples.shape[1:], self.seeds, subseeds=self.subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w) + noise = self.rng.next() # GC now before running the next img2img to prevent running out of memory devices.torch_gc() @@ -1418,7 +1349,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.image_conditioning = self.img2img_image_conditioning(image, self.init_latent, image_mask) def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): - x = create_random_tensors([opt_C, self.height // opt_f, self.width // opt_f], seeds=seeds, subseeds=subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self) + x = self.rng.next() if self.initial_noise_multiplier != 1.0: self.extra_generation_params["Noise multiplier"] = self.initial_noise_multiplier diff --git a/modules/rng.py b/modules/rng.py new file mode 100644 index 00000000..2d7baea5 --- /dev/null +++ b/modules/rng.py @@ -0,0 +1,171 @@ +import torch + +from modules import devices, rng_philox, shared + + +def randn(seed, shape, generator=None): + """Generate a tensor with random numbers from a normal distribution using seed. + + Uses the seed parameter to set the global torch seed; to generate more with that seed, use randn_like/randn_without_seed.""" + + manual_seed(seed) + + if shared.opts.randn_source == "NV": + return torch.asarray((generator or nv_rng).randn(shape), device=devices.device) + + if shared.opts.randn_source == "CPU" or devices.device.type == 'mps': + return torch.randn(shape, device=devices.cpu, generator=generator).to(devices.device) + + return torch.randn(shape, device=devices.device, generator=generator) + + +def randn_local(seed, shape): + """Generate a tensor with random numbers from a normal distribution using seed. + + Does not change the global random number generator. You can only generate the seed's first tensor using this function.""" + + if shared.opts.randn_source == "NV": + rng = rng_philox.Generator(seed) + return torch.asarray(rng.randn(shape), device=devices.device) + + local_device = devices.cpu if shared.opts.randn_source == "CPU" or devices.device.type == 'mps' else devices.device + local_generator = torch.Generator(local_device).manual_seed(int(seed)) + return torch.randn(shape, device=local_device, generator=local_generator).to(devices.device) + + +def randn_like(x): + """Generate a tensor with random numbers from a normal distribution using the previously initialized genrator. + + Use either randn() or manual_seed() to initialize the generator.""" + + if shared.opts.randn_source == "NV": + return torch.asarray(nv_rng.randn(x.shape), device=x.device, dtype=x.dtype) + + if shared.opts.randn_source == "CPU" or x.device.type == 'mps': + return torch.randn_like(x, device=devices.cpu).to(x.device) + + return torch.randn_like(x) + + +def randn_without_seed(shape, generator=None): + """Generate a tensor with random numbers from a normal distribution using the previously initialized genrator. + + Use either randn() or manual_seed() to initialize the generator.""" + + if shared.opts.randn_source == "NV": + return torch.asarray((generator or nv_rng).randn(shape), device=devices.device) + + if shared.opts.randn_source == "CPU" or devices.device.type == 'mps': + return torch.randn(shape, device=devices.cpu, generator=generator).to(devices.device) + + return torch.randn(shape, device=devices.device, generator=generator) + + +def manual_seed(seed): + """Set up a global random number generator using the specified seed.""" + from modules.shared import opts + + if opts.randn_source == "NV": + global nv_rng + nv_rng = rng_philox.Generator(seed) + return + + torch.manual_seed(seed) + + +def create_generator(seed): + if shared.opts.randn_source == "NV": + return rng_philox.Generator(seed) + + device = devices.cpu if shared.opts.randn_source == "CPU" or devices.device.type == 'mps' else devices.device + generator = torch.Generator(device).manual_seed(int(seed)) + return generator + + +# from https://discuss.pytorch.org/t/help-regarding-slerp-function-for-generative-model-sampling/32475/3 +def slerp(val, low, high): + low_norm = low/torch.norm(low, dim=1, keepdim=True) + high_norm = high/torch.norm(high, dim=1, keepdim=True) + dot = (low_norm*high_norm).sum(1) + + if dot.mean() > 0.9995: + return low * val + high * (1 - val) + + omega = torch.acos(dot) + so = torch.sin(omega) + res = (torch.sin((1.0-val)*omega)/so).unsqueeze(1)*low + (torch.sin(val*omega)/so).unsqueeze(1) * high + return res + + +class ImageRNG: + def __init__(self, shape, seeds, subseeds=None, subseed_strength=0.0, seed_resize_from_h=0, seed_resize_from_w=0): + self.shape = shape + self.seeds = seeds + self.subseeds = subseeds + self.subseed_strength = subseed_strength + self.seed_resize_from_h = seed_resize_from_h + self.seed_resize_from_w = seed_resize_from_w + + self.generators = [create_generator(seed) for seed in seeds] + + self.is_first = True + + def first(self): + noise_shape = self.shape if self.seed_resize_from_h <= 0 or self.seed_resize_from_w <= 0 else (self.shape[0], self.seed_resize_from_h // 8, self.seed_resize_from_w // 8) + + xs = [] + + for i, (seed, generator) in enumerate(zip(self.seeds, self.generators)): + subnoise = None + if self.subseeds is not None and self.subseed_strength != 0: + subseed = 0 if i >= len(self.subseeds) else self.subseeds[i] + subnoise = randn(subseed, noise_shape) + + if noise_shape != self.shape: + noise = randn(seed, noise_shape) + else: + noise = randn(seed, self.shape, generator=generator) + + if subnoise is not None: + noise = slerp(self.subseed_strength, noise, subnoise) + + if noise_shape != self.shape: + x = randn(seed, self.shape, generator=generator) + dx = (self.shape[2] - noise_shape[2]) // 2 + dy = (self.shape[1] - noise_shape[1]) // 2 + w = noise_shape[2] if dx >= 0 else noise_shape[2] + 2 * dx + h = noise_shape[1] if dy >= 0 else noise_shape[1] + 2 * dy + tx = 0 if dx < 0 else dx + ty = 0 if dy < 0 else dy + dx = max(-dx, 0) + dy = max(-dy, 0) + + x[:, ty:ty + h, tx:tx + w] = noise[:, dy:dy + h, dx:dx + w] + noise = x + + xs.append(noise) + + eta_noise_seed_delta = shared.opts.eta_noise_seed_delta or 0 + if eta_noise_seed_delta: + self.generators = [create_generator(seed + eta_noise_seed_delta) for seed in self.seeds] + + return torch.stack(xs).to(shared.device) + + def next(self): + if self.is_first: + self.is_first = False + return self.first() + + xs = [] + for generator in self.generators: + x = randn_without_seed(self.shape, generator=generator) + xs.append(x) + + return torch.stack(xs).to(shared.device) + + +devices.randn = randn +devices.randn_local = randn_local +devices.randn_like = randn_like +devices.randn_without_seed = randn_without_seed +devices.manual_seed = manual_seed diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index adda963b..97bc0804 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -1,5 +1,5 @@ import inspect -from collections import namedtuple, deque +from collections import namedtuple import numpy as np import torch from PIL import Image @@ -132,10 +132,15 @@ replace_torchsde_browinan() class TorchHijack: - def __init__(self, sampler_noises): - # Using a deque to efficiently receive the sampler_noises in the same order as the previous index-based - # implementation. - self.sampler_noises = deque(sampler_noises) + """This is here to replace torch.randn_like of k-diffusion. + + k-diffusion has random_sampler argument for most samplers, but not for all, so + this is needed to properly replace every use of torch.randn_like. + + We need to replace to make images generated in batches to be same as images generated individually.""" + + def __init__(self, p): + self.rng = p.rng def __getattr__(self, item): if item == 'randn_like': @@ -147,12 +152,7 @@ class TorchHijack: raise AttributeError(f"'{type(self).__name__}' object has no attribute '{item}'") def randn_like(self, x): - if self.sampler_noises: - noise = self.sampler_noises.popleft() - if noise.shape == x.shape: - return noise - - return devices.randn_like(x) + return self.rng.next() class Sampler: @@ -215,7 +215,7 @@ class Sampler: self.eta = p.eta if p.eta is not None else getattr(opts, self.eta_option_field, 0.0) self.s_min_uncond = getattr(p, 's_min_uncond', 0.0) - k_diffusion.sampling.torch = TorchHijack(self.sampler_noises if self.sampler_noises is not None else []) + k_diffusion.sampling.torch = TorchHijack(p) extra_params_kwargs = {} for param_name in self.extra_params: diff --git a/modules/shared.py b/modules/shared.py index e34847ce..e9b980a4 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -16,7 +16,7 @@ import modules.interrogate import modules.memmon import modules.styles import modules.devices as devices -from modules import localization, script_loading, errors, ui_components, shared_items, cmd_args +from modules import localization, script_loading, errors, ui_components, shared_items, cmd_args, rng # noqa: F401 from modules.paths_internal import models_path, script_path, data_path, sd_configs_path, sd_default_config, sd_model_file, default_sd_model_file, extensions_dir, extensions_builtin_dir # noqa: F401 from ldm.models.diffusion.ddpm import LatentDiffusion from typing import Optional -- cgit v1.2.3 From d1ba46b6e15d2f2917e7d392955c8c6f988527ba Mon Sep 17 00:00:00 2001 From: Robert Barron Date: Wed, 9 Aug 2023 07:46:30 -0700 Subject: allow first pass and hires pass to use a single prompt to do different prompt editing, hires is 1.0..2.0: relative time range is [1..2] absolute time range is [steps+1..steps+hire_steps], e.g. with 30 steps and 20 hires steps, '20' is 2/3rds through first pass, and 40 is halfway through hires pass --- modules/processing.py | 20 ++++++++++++-------- modules/prompt_parser.py | 41 +++++++++++++++++++++++++++++++---------- modules/shared.py | 1 + 3 files changed, 44 insertions(+), 18 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 31745006..0750b299 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -319,12 +319,14 @@ class StableDiffusionProcessing: self.all_prompts = [shared.prompt_styles.apply_styles_to_prompt(x, self.styles) for x in self.all_prompts] self.all_negative_prompts = [shared.prompt_styles.apply_negative_styles_to_prompt(x, self.styles) for x in self.all_negative_prompts] - def cached_params(self, required_prompts, steps, extra_network_data): + def cached_params(self, required_prompts, steps, hires_steps, extra_network_data, use_old_scheduling): """Returns parameters that invalidate the cond cache if changed""" return ( required_prompts, steps, + hires_steps, + use_old_scheduling, opts.CLIP_stop_at_last_layers, shared.sd_model.sd_checkpoint_info, extra_network_data, @@ -334,7 +336,7 @@ class StableDiffusionProcessing: self.height, ) - def get_conds_with_caching(self, function, required_prompts, steps, caches, extra_network_data): + def get_conds_with_caching(self, function, required_prompts, steps, hires_steps, caches, extra_network_data): """ Returns the result of calling function(shared.sd_model, required_prompts, steps) using a cache to store the result if the same arguments have been used before. @@ -347,7 +349,7 @@ class StableDiffusionProcessing: caches is a list with items described above. """ - cached_params = self.cached_params(required_prompts, steps, extra_network_data) + cached_params = self.cached_params(required_prompts, steps, hires_steps, extra_network_data, shared.opts.use_old_scheduling) for cache in caches: if cache[0] is not None and cached_params == cache[0]: @@ -356,7 +358,7 @@ class StableDiffusionProcessing: cache = caches[0] with devices.autocast(): - cache[1] = function(shared.sd_model, required_prompts, steps) + cache[1] = function(shared.sd_model, required_prompts, steps, hires_steps, shared.opts.use_old_scheduling) cache[0] = cached_params return cache[1] @@ -367,8 +369,9 @@ class StableDiffusionProcessing: sampler_config = sd_samplers.find_sampler_config(self.sampler_name) self.step_multiplier = 2 if sampler_config and sampler_config.options.get("second_order", False) else 1 - self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, self.steps * self.step_multiplier, [self.cached_uc], self.extra_network_data) - self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, self.steps * self.step_multiplier, [self.cached_c], self.extra_network_data) + self.firstpass_steps = self.steps * self.step_multiplier + self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, self.firstpass_steps, None, [self.cached_uc], self.extra_network_data) + self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, self.firstpass_steps, None, [self.cached_c], self.extra_network_data) def parse_extra_network_prompts(self): self.prompts, self.extra_network_data = extra_networks.parse_prompts(self.prompts) @@ -1225,8 +1228,9 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): hr_prompts = prompt_parser.SdConditioning(self.hr_prompts, width=self.hr_upscale_to_x, height=self.hr_upscale_to_y) hr_negative_prompts = prompt_parser.SdConditioning(self.hr_negative_prompts, width=self.hr_upscale_to_x, height=self.hr_upscale_to_y, is_negative_prompt=True) - self.hr_uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, hr_negative_prompts, self.steps * self.step_multiplier, [self.cached_hr_uc, self.cached_uc], self.hr_extra_network_data) - self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, hr_prompts, self.steps * self.step_multiplier, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) + hires_steps = (self.hr_second_pass_steps or self.steps) * self.step_multiplier + self.hr_uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, hr_negative_prompts, self.firstpass_steps, hires_steps, [self.cached_hr_uc, self.cached_uc], self.hr_extra_network_data) + self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, hr_prompts, self.firstpass_steps, hires_steps, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) def setup_conds(self): super().setup_conds() diff --git a/modules/prompt_parser.py b/modules/prompt_parser.py index 32d214e3..e8c41f38 100644 --- a/modules/prompt_parser.py +++ b/modules/prompt_parser.py @@ -26,7 +26,7 @@ plain: /([^\\\[\]():|]|\\.)+/ %import common.SIGNED_NUMBER -> NUMBER """) -def get_learned_conditioning_prompt_schedules(prompts, steps): +def get_learned_conditioning_prompt_schedules(prompts, base_steps, hires_steps=None, use_old_scheduling=False): """ >>> g = lambda p: get_learned_conditioning_prompt_schedules([p], 10)[0] >>> g("test") @@ -57,18 +57,39 @@ def get_learned_conditioning_prompt_schedules(prompts, steps): [[1, 'female'], [2, 'male'], [3, 'female'], [4, 'male'], [5, 'female'], [6, 'male'], [7, 'female'], [8, 'male'], [9, 'female'], [10, 'male']] >>> g("[fe|||]male") [[1, 'female'], [2, 'male'], [3, 'male'], [4, 'male'], [5, 'female'], [6, 'male'], [7, 'male'], [8, 'male'], [9, 'female'], [10, 'male']] + >>> g = lambda p: get_learned_conditioning_prompt_schedules([p], 10, 10)[0] + >>> g("a [b:.5] c") + [[10, 'a b c']] + >>> g("a [b:1.5] c") + [[5, 'a c'], [10, 'a b c']] """ + if hires_steps is None or use_old_scheduling: + int_offset = 0 + flt_offset = 0 + steps = base_steps + else: + int_offset = base_steps + flt_offset = 1.0 + steps = hires_steps + def collect_steps(steps, tree): res = [steps] class CollectSteps(lark.Visitor): def scheduled(self, tree): - tree.children[-2] = float(tree.children[-2]) - if tree.children[-2] < 1: - tree.children[-2] *= steps - tree.children[-2] = min(steps, int(tree.children[-2])) - res.append(tree.children[-2]) + s = tree.children[-2] + v = float(s) + if use_old_scheduling: + v = v*steps if v<1 else v + else: + if "." in s: + v = (v - flt_offset) * steps + else: + v = (v - int_offset) + tree.children[-2] = min(steps, int(v)) + if tree.children[-2] >= 1: + res.append(tree.children[-2]) def alternate(self, tree): res.extend(range(1, steps+1)) @@ -134,7 +155,7 @@ class SdConditioning(list): -def get_learned_conditioning(model, prompts: SdConditioning | list[str], steps): +def get_learned_conditioning(model, prompts: SdConditioning | list[str], steps, hires_steps=None, use_old_scheduling=False): """converts a list of prompts into a list of prompt schedules - each schedule is a list of ScheduledPromptConditioning, specifying the comdition (cond), and the sampling step at which this condition is to be replaced by the next one. @@ -154,7 +175,7 @@ def get_learned_conditioning(model, prompts: SdConditioning | list[str], steps): """ res = [] - prompt_schedules = get_learned_conditioning_prompt_schedules(prompts, steps) + prompt_schedules = get_learned_conditioning_prompt_schedules(prompts, steps, hires_steps, use_old_scheduling) cache = {} for prompt, prompt_schedule in zip(prompts, prompt_schedules): @@ -229,7 +250,7 @@ class MulticondLearnedConditioning: self.batch: List[List[ComposableScheduledPromptConditioning]] = batch -def get_multicond_learned_conditioning(model, prompts, steps) -> MulticondLearnedConditioning: +def get_multicond_learned_conditioning(model, prompts, steps, hires_steps=None, use_old_scheduling=False) -> MulticondLearnedConditioning: """same as get_learned_conditioning, but returns a list of ScheduledPromptConditioning along with the weight objects for each prompt. For each prompt, the list is obtained by splitting the prompt using the AND separator. @@ -238,7 +259,7 @@ def get_multicond_learned_conditioning(model, prompts, steps) -> MulticondLearne res_indexes, prompt_flat_list, prompt_indexes = get_multicond_prompt_list(prompts) - learned_conditioning = get_learned_conditioning(model, prompt_flat_list, steps) + learned_conditioning = get_learned_conditioning(model, prompt_flat_list, steps, hires_steps, use_old_scheduling) res = [] for indexes in res_indexes: diff --git a/modules/shared.py b/modules/shared.py index 078e8135..a605b08b 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -516,6 +516,7 @@ options_templates.update(options_section(('compatibility', "Compatibility"), { "use_old_hires_fix_width_height": OptionInfo(False, "For hires fix, use width/height sliders to set final resolution rather than first pass (disables Upscale by, Resize width/height to)."), "dont_fix_second_order_samplers_schedule": OptionInfo(False, "Do not fix prompt schedule for second order samplers."), "hires_fix_use_firstpass_conds": OptionInfo(False, "For hires fix, calculate conds of second pass using extra networks of first pass."), + "use_old_scheduling": OptionInfo(False, "Use old prompt where first pass and hires both used the same timeline, and < 1 meant relative and >= 1 meant absolute"), })) options_templates.update(options_section(('interrogate', "Interrogate"), { -- cgit v1.2.3 From 259805947e4e567abcf6a4be20fe520d456c251f Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Wed, 9 Aug 2023 14:24:16 -0400 Subject: Add slerp import for extension backwards compat --- modules/processing.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 2df5e8c7..6961b7b1 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -15,6 +15,7 @@ from typing import Any, Dict, List import modules.sd_hijack from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng +from modules.rng import slerp # noqa: F401 from modules.sd_hijack import model_hijack from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes from modules.shared import opts, cmd_opts, state -- cgit v1.2.3 From 9199b6b7ebe96cdf09571ba874a103e8ed8c90ef Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 10 Aug 2023 11:20:46 +0300 Subject: add a custom UI element that combines accordion and checkbox rework hires fix UI to use accordion prevent bogus progress output in console when calculating hires fix dimensions --- javascript/inputAccordion.js | 37 +++++++++++++++++++++ modules/processing.py | 77 +++++++++++++++++++++++--------------------- modules/ui.py | 23 ++++--------- modules/ui_components.py | 31 ++++++++++++++++++ style.css | 15 +++++---- 5 files changed, 124 insertions(+), 59 deletions(-) create mode 100644 javascript/inputAccordion.js (limited to 'modules/processing.py') diff --git a/javascript/inputAccordion.js b/javascript/inputAccordion.js new file mode 100644 index 00000000..a5eef229 --- /dev/null +++ b/javascript/inputAccordion.js @@ -0,0 +1,37 @@ +var observerAccordionOpen = new MutationObserver(function(mutations) { + mutations.forEach(function(mutationRecord) { + var elem = mutationRecord.target; + var open = elem.classList.contains('open'); + + var accordion = elem.parentNode; + accordion.classList.toggle('input-accordion-open', open); + + var checkbox = gradioApp().querySelector('#' + accordion.id + "-checkbox input"); + checkbox.checked = open; + updateInput(checkbox); + + extra = gradioApp().querySelector('#' + accordion.id + "-extra"); + if(extra){ + extra.style.display = open ? "" : "none"; + } + }); +}); + +function inputAccordionChecked(id, checked){ + var label = gradioApp().querySelector('#' + id + " .label-wrap"); + if(label.classList.contains('open') != checked){ + label.click(); + } +} + +onUiLoaded(function() { + for (var accordion of gradioApp().querySelectorAll('.input-accordion')) { + var labelWrap = accordion.querySelector('.label-wrap'); + observerAccordionOpen.observe(labelWrap, {attributes: true, attributeFilter: ['class']}); + + var extra = gradioApp().querySelector('#' + accordion.id + "-extra"); + if(extra){ + labelWrap.insertBefore(extra, labelWrap.lastElementChild) + } + } +}); diff --git a/modules/processing.py b/modules/processing.py index 6961b7b1..7819644c 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -924,6 +924,45 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): self.hr_c = None self.hr_uc = None + def calculate_target_resolution(self): + if opts.use_old_hires_fix_width_height and self.applied_old_hires_behavior_to != (self.width, self.height): + self.hr_resize_x = self.width + self.hr_resize_y = self.height + self.hr_upscale_to_x = self.width + self.hr_upscale_to_y = self.height + + self.width, self.height = old_hires_fix_first_pass_dimensions(self.width, self.height) + self.applied_old_hires_behavior_to = (self.width, self.height) + + if self.hr_resize_x == 0 and self.hr_resize_y == 0: + self.extra_generation_params["Hires upscale"] = self.hr_scale + self.hr_upscale_to_x = int(self.width * self.hr_scale) + self.hr_upscale_to_y = int(self.height * self.hr_scale) + else: + self.extra_generation_params["Hires resize"] = f"{self.hr_resize_x}x{self.hr_resize_y}" + + if self.hr_resize_y == 0: + self.hr_upscale_to_x = self.hr_resize_x + self.hr_upscale_to_y = self.hr_resize_x * self.height // self.width + elif self.hr_resize_x == 0: + self.hr_upscale_to_x = self.hr_resize_y * self.width // self.height + self.hr_upscale_to_y = self.hr_resize_y + else: + target_w = self.hr_resize_x + target_h = self.hr_resize_y + src_ratio = self.width / self.height + dst_ratio = self.hr_resize_x / self.hr_resize_y + + if src_ratio < dst_ratio: + self.hr_upscale_to_x = self.hr_resize_x + self.hr_upscale_to_y = self.hr_resize_x * self.height // self.width + else: + self.hr_upscale_to_x = self.hr_resize_y * self.width // self.height + self.hr_upscale_to_y = self.hr_resize_y + + self.truncate_x = (self.hr_upscale_to_x - target_w) // opt_f + self.truncate_y = (self.hr_upscale_to_y - target_h) // opt_f + def init(self, all_prompts, all_seeds, all_subseeds): if self.enable_hr: if self.hr_checkpoint_name: @@ -948,43 +987,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): if not any(x.name == self.hr_upscaler for x in shared.sd_upscalers): raise Exception(f"could not find upscaler named {self.hr_upscaler}") - if opts.use_old_hires_fix_width_height and self.applied_old_hires_behavior_to != (self.width, self.height): - self.hr_resize_x = self.width - self.hr_resize_y = self.height - self.hr_upscale_to_x = self.width - self.hr_upscale_to_y = self.height - - self.width, self.height = old_hires_fix_first_pass_dimensions(self.width, self.height) - self.applied_old_hires_behavior_to = (self.width, self.height) - - if self.hr_resize_x == 0 and self.hr_resize_y == 0: - self.extra_generation_params["Hires upscale"] = self.hr_scale - self.hr_upscale_to_x = int(self.width * self.hr_scale) - self.hr_upscale_to_y = int(self.height * self.hr_scale) - else: - self.extra_generation_params["Hires resize"] = f"{self.hr_resize_x}x{self.hr_resize_y}" - - if self.hr_resize_y == 0: - self.hr_upscale_to_x = self.hr_resize_x - self.hr_upscale_to_y = self.hr_resize_x * self.height // self.width - elif self.hr_resize_x == 0: - self.hr_upscale_to_x = self.hr_resize_y * self.width // self.height - self.hr_upscale_to_y = self.hr_resize_y - else: - target_w = self.hr_resize_x - target_h = self.hr_resize_y - src_ratio = self.width / self.height - dst_ratio = self.hr_resize_x / self.hr_resize_y - - if src_ratio < dst_ratio: - self.hr_upscale_to_x = self.hr_resize_x - self.hr_upscale_to_y = self.hr_resize_x * self.height // self.width - else: - self.hr_upscale_to_x = self.hr_resize_y * self.width // self.height - self.hr_upscale_to_y = self.hr_resize_y - - self.truncate_x = (self.hr_upscale_to_x - target_w) // opt_f - self.truncate_y = (self.hr_upscale_to_y - target_h) // opt_f + self.calculate_target_resolution() if not state.processing_has_refined_job_count: if state.job_count == -1: diff --git a/modules/ui.py b/modules/ui.py index 4e1daa8d..cbad3afe 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -14,7 +14,7 @@ from modules.call_queue import wrap_gradio_gpu_call, wrap_queued_call, wrap_grad from modules import gradio_extensons # noqa: F401 from modules import sd_hijack, sd_models, script_callbacks, ui_extensions, deepbooru, extra_networks, ui_common, ui_postprocessing, progress, ui_loadsave, errors, shared_items, ui_settings, timer, sysinfo, ui_checkpoint_merger, ui_prompt_styles, scripts, sd_samplers, processing, devices, ui_extra_networks -from modules.ui_components import FormRow, FormGroup, ToolButton, FormHTML +from modules.ui_components import FormRow, FormGroup, ToolButton, FormHTML, InputAccordion from modules.paths import script_path from modules.ui_common import create_refresh_button from modules.ui_gradio_extensions import reload_javascript @@ -94,11 +94,9 @@ def calc_resolution_hires(enable, width, height, hr_scale, hr_resize_x, hr_resiz return "" p = processing.StableDiffusionProcessingTxt2Img(width=width, height=height, enable_hr=True, hr_scale=hr_scale, hr_resize_x=hr_resize_x, hr_resize_y=hr_resize_y) + p.calculate_target_resolution() - with devices.autocast(): - p.init([""], [0], [0]) - - return f"resize: from {p.width}x{p.height} to {p.hr_resize_x or p.hr_upscale_to_x}x{p.hr_resize_y or p.hr_upscale_to_y}" + return f"from {p.width}x{p.height} to {p.hr_resize_x or p.hr_upscale_to_x}x{p.hr_resize_y or p.hr_upscale_to_y}" def resize_from_to_html(width, height, scale_by): @@ -436,11 +434,12 @@ def create_ui(): with FormRow(elem_classes="checkboxes-row", variant="compact"): restore_faces = gr.Checkbox(label='Restore faces', value=False, visible=len(shared.face_restorers) > 1, elem_id="txt2img_restore_faces") tiling = gr.Checkbox(label='Tiling', value=False, elem_id="txt2img_tiling") - enable_hr = gr.Checkbox(label='Hires. fix', value=False, elem_id="txt2img_enable_hr") - hr_final_resolution = FormHTML(value="", elem_id="txtimg_hr_finalres", label="Upscaled resolution", interactive=False) elif category == "hires_fix": - with FormGroup(visible=False, elem_id="txt2img_hires_fix") as hr_options: + with InputAccordion(False, label="Hires. fix") as enable_hr: + with enable_hr.extra(): + hr_final_resolution = FormHTML(value="", elem_id="txtimg_hr_finalres", label="Upscaled resolution", interactive=False, min_width=0) + with FormRow(elem_id="txt2img_hires_fix_row1", variant="compact"): hr_upscaler = gr.Dropdown(label="Upscaler", elem_id="txt2img_hr_upscaler", choices=[*shared.latent_upscale_modes, *[x.name for x in shared.sd_upscalers]], value=shared.latent_upscale_default_mode) hr_second_pass_steps = gr.Slider(minimum=0, maximum=150, step=1, label='Hires steps', value=0, elem_id="txt2img_hires_steps") @@ -568,13 +567,6 @@ def create_ui(): show_progress=False, ) - enable_hr.change( - fn=lambda x: gr_show(x), - inputs=[enable_hr], - outputs=[hr_options], - show_progress = False, - ) - txt2img_paste_fields = [ (toprow.prompt, "Prompt"), (toprow.negative_prompt, "Negative prompt"), @@ -594,7 +586,6 @@ def create_ui(): (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), (denoising_strength, "Denoising strength"), (enable_hr, lambda d: "Denoising strength" in d and ("Hires upscale" in d or "Hires upscaler" in d or "Hires resize-1" in d)), - (hr_options, lambda d: gr.Row.update(visible="Denoising strength" in d and ("Hires upscale" in d or "Hires upscaler" in d or "Hires resize-1" in d))), (hr_scale, "Hires upscale"), (hr_upscaler, "Hires upscaler"), (hr_second_pass_steps, "Hires steps"), diff --git a/modules/ui_components.py b/modules/ui_components.py index 8f8a7088..598ce738 100644 --- a/modules/ui_components.py +++ b/modules/ui_components.py @@ -72,3 +72,34 @@ class DropdownEditable(FormComponent, gr.Dropdown): def get_block_name(self): return "dropdown" + +class InputAccordion(gr.Checkbox): + global_index = 0 + + def __init__(self, value, **kwargs): + self.accordion_id = kwargs.get('elem_id') + if self.accordion_id is None: + self.accordion_id = f"input-accordion-{self.global_index}" + self.global_index += 1 + + kwargs['elem_id'] = self.accordion_id + "-checkbox" + kwargs['visible'] = False + super().__init__(value, **kwargs) + + self.change(fn=None, _js='function(checked){ inputAccordionChecked("' + self.accordion_id + '", checked); }', inputs=[self]) + + self.accordion = gr.Accordion(kwargs.get('label', 'Accordion'), open=value, elem_id=self.accordion_id, elem_classes=['input-accordion']) + + def extra(self): + return gr.Column(elem_id=self.accordion_id + '-extra', elem_classes='input-accordion-extra', min_width=0) + + def __enter__(self): + self.accordion.__enter__() + return self + + def __exit__(self, exc_type, exc_val, exc_tb): + self.accordion.__exit__(exc_type, exc_val, exc_tb) + + def get_block_name(self): + return "checkbox" + diff --git a/style.css b/style.css index dfd5ca36..5163e53c 100644 --- a/style.css +++ b/style.css @@ -329,12 +329,6 @@ div.gradio-accordion { border-radius: 0 0.5rem 0.5rem 0; } -#txtimg_hr_finalres{ - min-height: 0 !important; - padding: .625rem .75rem; - margin-left: -0.75em -} - #img2img_scale_resolution_preview.block{ display: flex; align-items: end; @@ -1016,3 +1010,12 @@ div.block.gradio-box.popup-dialog, .popup-dialog { div.block.gradio-box.popup-dialog > div:last-child, .popup-dialog > div:last-child{ margin-top: 1em; } + +div.block.input-accordion{ + margin-bottom: 0.4em; +} + +.input-accordion-extra{ + flex: 0 0 auto !important; + margin: 0 0.5em 0 auto; +} -- cgit v1.2.3 From 33446acf47a8c3e0c0964782189562df3c4bcf4f Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 10 Aug 2023 12:41:41 +0300 Subject: face restoration and tiling moved to settings - use "Options in main UI" setting if you want them back --- modules/generation_parameters_copypaste.py | 2 ++ modules/img2img.py | 4 +--- modules/processing.py | 11 +++++++++-- modules/shared_options.py | 2 ++ modules/txt2img.py | 4 +--- modules/ui.py | 12 ++---------- 6 files changed, 17 insertions(+), 18 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index d932c67d..bdff3266 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -343,6 +343,8 @@ infotext_to_setting_name_mapping = [ ('Pad conds', 'pad_cond_uncond'), ('VAE Encoder', 'sd_vae_encode_method'), ('VAE Decoder', 'sd_vae_decode_method'), + ('Tiling', 'tiling'), + ('Face restoration', 'face_restoration'), ] diff --git a/modules/img2img.py b/modules/img2img.py index e06ac1d6..c7bbbac8 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -116,7 +116,7 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal process_images(p) -def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_name: str, mask_blur: int, mask_alpha: float, inpainting_fill: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, img2img_batch_use_png_info: bool, img2img_batch_png_info_props: list, img2img_batch_png_info_dir: str, request: gr.Request, *args): +def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_name: str, mask_blur: int, mask_alpha: float, inpainting_fill: int, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, img2img_batch_use_png_info: bool, img2img_batch_png_info_props: list, img2img_batch_png_info_dir: str, request: gr.Request, *args): override_settings = create_override_settings_dict(override_settings_texts) is_batch = mode == 5 @@ -179,8 +179,6 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s cfg_scale=cfg_scale, width=width, height=height, - restore_faces=restore_faces, - tiling=tiling, init_images=[image], mask=mask, mask_blur=mask_blur, diff --git a/modules/processing.py b/modules/processing.py index 7819644c..68a8f1c6 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -111,7 +111,7 @@ class StableDiffusionProcessing: cached_uc = [None, None] cached_c = [None, None] - def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = False, tiling: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = None, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): + def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = None, tiling: bool = None, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = None, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): if sampler_index is not None: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) @@ -564,7 +564,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "CFG scale": p.cfg_scale, "Image CFG scale": getattr(p, 'image_cfg_scale', None), "Seed": p.all_seeds[0] if use_main_prompt else all_seeds[index], - "Face restoration": (opts.face_restoration_model if p.restore_faces else None), + "Face restoration": opts.face_restoration_model if p.restore_faces else None, "Size": f"{p.width}x{p.height}", "Model hash": getattr(p, 'sd_model_hash', None if not opts.add_model_hash_to_info or not shared.sd_model.sd_model_hash else shared.sd_model.sd_model_hash), "Model": (None if not opts.add_model_name_to_info else shared.sd_model.sd_checkpoint_info.name_for_extra), @@ -580,6 +580,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Init image hash": getattr(p, 'init_img_hash', None), "RNG": opts.randn_source if opts.randn_source != "GPU" and opts.randn_source != "NV" else None, "NGMS": None if p.s_min_uncond == 0 else p.s_min_uncond, + "Tiling": "True" if p.tiling else None, **p.extra_generation_params, "Version": program_version() if opts.add_version_to_infotext else None, "User": p.user if opts.add_user_name_to_info else None, @@ -645,6 +646,12 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: seed = get_fixed_seed(p.seed) subseed = get_fixed_seed(p.subseed) + if p.restore_faces is None: + p.restore_faces = opts.face_restoration + + if p.tiling is None: + p.tiling = opts.tiling + modules.sd_hijack.model_hijack.apply_circular(p.tiling) modules.sd_hijack.model_hijack.clear_comments() diff --git a/modules/shared_options.py b/modules/shared_options.py index 7468bc81..f72859d9 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -92,6 +92,7 @@ options_templates.update(options_section(('upscaling', "Upscaling"), { })) options_templates.update(options_section(('face-restoration', "Face restoration"), { + "face_restoration": OptionInfo(False, "Restore faces").info("will use a third-party model on generation result to reconstruct faces"), "face_restoration_model": OptionInfo("CodeFormer", "Face restoration model", gr.Radio, lambda: {"choices": [x.name() for x in shared.face_restorers]}), "code_former_weight": OptionInfo(0.5, "CodeFormer weight", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01}).info("0 = maximum effect; 1 = minimum effect"), "face_restoration_unload": OptionInfo(False, "Move face restoration model from VRAM into RAM after processing"), @@ -138,6 +139,7 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "CLIP_stop_at_last_layers": OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}).link("wiki", "https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#clip-skip").info("ignore last layers of CLIP network; 1 ignores none, 2 ignores one layer"), "upcast_attn": OptionInfo(False, "Upcast cross attention layer to float32"), "randn_source": OptionInfo("GPU", "Random number generator source.", gr.Radio, {"choices": ["GPU", "CPU", "NV"]}).info("changes seeds drastically; use CPU to produce the same picture across different videocard vendors; use NV to produce same picture as on NVidia videocards"), + "tiling": OptionInfo(False, "Tiling").info("produce a tileable picture"), })) options_templates.update(options_section(('sdxl', "Stable Diffusion XL"), { diff --git a/modules/txt2img.py b/modules/txt2img.py index edad8930..5ea96bba 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -9,7 +9,7 @@ from modules.ui import plaintext_to_html import gradio as gr -def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_name: str, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_name: str, hr_prompt: str, hr_negative_prompt, override_settings_texts, request: gr.Request, *args): +def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_name: str, n_iter: int, batch_size: int, cfg_scale: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_name: str, hr_prompt: str, hr_negative_prompt, override_settings_texts, request: gr.Request, *args): override_settings = create_override_settings_dict(override_settings_texts) p = processing.StableDiffusionProcessingTxt2Img( @@ -32,8 +32,6 @@ def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, step cfg_scale=cfg_scale, width=width, height=height, - restore_faces=restore_faces, - tiling=tiling, enable_hr=enable_hr, denoising_strength=denoising_strength if enable_hr else None, hr_scale=hr_scale, diff --git a/modules/ui.py b/modules/ui.py index cbad3afe..09a826fd 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -432,8 +432,7 @@ def create_ui(): elif category == "checkboxes": with FormRow(elem_classes="checkboxes-row", variant="compact"): - restore_faces = gr.Checkbox(label='Restore faces', value=False, visible=len(shared.face_restorers) > 1, elem_id="txt2img_restore_faces") - tiling = gr.Checkbox(label='Tiling', value=False, elem_id="txt2img_tiling") + pass elif category == "hires_fix": with InputAccordion(False, label="Hires. fix") as enable_hr: @@ -516,8 +515,6 @@ def create_ui(): toprow.ui_styles.dropdown, steps, sampler_name, - restore_faces, - tiling, batch_count, batch_size, cfg_scale, @@ -572,7 +569,6 @@ def create_ui(): (toprow.negative_prompt, "Negative prompt"), (steps, "Steps"), (sampler_name, "Sampler"), - (restore_faces, "Face restoration"), (cfg_scale, "CFG scale"), (seed, "Seed"), (width, "Size-1"), @@ -792,8 +788,7 @@ def create_ui(): elif category == "checkboxes": with FormRow(elem_classes="checkboxes-row", variant="compact"): - restore_faces = gr.Checkbox(label='Restore faces', value=False, visible=len(shared.face_restorers) > 1, elem_id="img2img_restore_faces") - tiling = gr.Checkbox(label='Tiling', value=False, elem_id="img2img_tiling") + pass elif category == "batch": if not opts.dimensions_and_batch_together: @@ -866,8 +861,6 @@ def create_ui(): mask_blur, mask_alpha, inpainting_fill, - restore_faces, - tiling, batch_count, batch_size, cfg_scale, @@ -959,7 +952,6 @@ def create_ui(): (toprow.negative_prompt, "Negative prompt"), (steps, "Steps"), (sampler_name, "Sampler"), - (restore_faces, "Face restoration"), (cfg_scale, "CFG scale"), (image_cfg_scale, "Image CFG scale"), (seed, "Seed"), -- cgit v1.2.3 From 1b3093fe3aedb20aa8d505ceeea7900ac592e6fe Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 10 Aug 2023 15:58:53 +0300 Subject: fix --use-textbox-seed --- modules/processing.py | 12 ++++++++++-- modules/ui.py | 6 +++++- modules/ui_loadsave.py | 5 ++++- 3 files changed, 19 insertions(+), 4 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 68a8f1c6..f06c374a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -521,7 +521,15 @@ def decode_latent_batch(model, batch, target_device=None, check_for_nans=False): def get_fixed_seed(seed): - if seed is None or seed == '' or seed == -1: + if seed == '' or seed is None: + seed = -1 + elif isinstance(seed, str): + try: + seed = int(seed) + except Exception: + seed = -1 + + if seed == -1: return int(random.randrange(4294967294)) return seed @@ -728,7 +736,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: # strength, which is saved as "Model Strength: 1.0" in the infotext if n == 0: with open(os.path.join(paths.data_path, "params.txt"), "w", encoding="utf8") as file: - processed = Processed(p, [], p.seed, "") + processed = Processed(p, []) file.write(processed.infotext(p, 0)) p.setup_conds() diff --git a/modules/ui.py b/modules/ui.py index b87e95a6..e7433cbd 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -144,7 +144,11 @@ def interrogate_deepbooru(image): def create_seed_inputs(target_interface): with FormRow(elem_id=f"{target_interface}_seed_row", variant="compact"): - seed = (gr.Textbox if cmd_opts.use_textbox_seed else gr.Number)(label='Seed', value=-1, elem_id=f"{target_interface}_seed") + if cmd_opts.use_textbox_seed: + seed = gr.Textbox(label='Seed', value="", elem_id=f"{target_interface}_seed") + else: + seed = gr.Number(label='Seed', value=-1, elem_id=f"{target_interface}_seed") + random_seed = ToolButton(random_symbol, elem_id=f"{target_interface}_random_seed", label='Random seed') reuse_seed = ToolButton(reuse_symbol, elem_id=f"{target_interface}_reuse_seed", label='Reuse seed') diff --git a/modules/ui_loadsave.py b/modules/ui_loadsave.py index 0052a5cc..99d763e1 100644 --- a/modules/ui_loadsave.py +++ b/modules/ui_loadsave.py @@ -8,7 +8,7 @@ from modules.ui_components import ToolButton class UiLoadsave: - """allows saving and restorig default values for gradio components""" + """allows saving and restoring default values for gradio components""" def __init__(self, filename): self.filename = filename @@ -48,6 +48,9 @@ class UiLoadsave: elif condition and not condition(saved_value): pass else: + if isinstance(x, gr.Textbox) and field == 'value': # due to an undersirable behavior of gr.Textbox, if you give it an int value instead of str, everything dies + saved_value = str(saved_value) + setattr(obj, field, saved_value) if init_field is not None: init_field(saved_value) -- cgit v1.2.3 From 9d78d317ae492db59ebf8b31fda9a049f6c9bd14 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 10 Aug 2023 16:22:10 +0300 Subject: add VAE to infotext --- modules/generation_parameters_copypaste.py | 1 + modules/processing.py | 2 ++ modules/sd_vae.py | 16 +++++++++++++++- 3 files changed, 18 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index bdff3266..6ace29cf 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -341,6 +341,7 @@ infotext_to_setting_name_mapping = [ ('RNG', 'randn_source'), ('NGMS', 's_min_uncond'), ('Pad conds', 'pad_cond_uncond'), + ('VAE', 'sd_vae'), ('VAE Encoder', 'sd_vae_encode_method'), ('VAE Decoder', 'sd_vae_decode_method'), ('Tiling', 'tiling'), diff --git a/modules/processing.py b/modules/processing.py index f06c374a..44d47e8c 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -576,6 +576,8 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Size": f"{p.width}x{p.height}", "Model hash": getattr(p, 'sd_model_hash', None if not opts.add_model_hash_to_info or not shared.sd_model.sd_model_hash else shared.sd_model.sd_model_hash), "Model": (None if not opts.add_model_name_to_info else shared.sd_model.sd_checkpoint_info.name_for_extra), + "VAE hash": sd_vae.get_loaded_vae_hash() if opts.add_model_hash_to_info else None, + "VAE": sd_vae.get_loaded_vae_name() if opts.add_model_name_to_info else None, "Variation seed": (None if p.subseed_strength == 0 else (p.all_subseeds[0] if use_main_prompt else all_subseeds[index])), "Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength), "Seed resize from": (None if p.seed_resize_from_w <= 0 or p.seed_resize_from_h <= 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"), diff --git a/modules/sd_vae.py b/modules/sd_vae.py index 5ac1ac31..1db01992 100644 --- a/modules/sd_vae.py +++ b/modules/sd_vae.py @@ -2,7 +2,7 @@ import os import collections from dataclasses import dataclass -from modules import paths, shared, devices, script_callbacks, sd_models, extra_networks, lowvram, sd_hijack +from modules import paths, shared, devices, script_callbacks, sd_models, extra_networks, lowvram, sd_hijack, hashes import glob from copy import deepcopy @@ -20,6 +20,20 @@ checkpoint_info = None checkpoints_loaded = collections.OrderedDict() +def get_loaded_vae_name(): + if loaded_vae_file is None: + return None + + return os.path.basename(loaded_vae_file) + + +def get_loaded_vae_hash(): + if loaded_vae_file is None: + return None + + return hashes.sha256(loaded_vae_file, 'vae')[0:10] + + def get_base_vae(model): if base_vae is not None and checkpoint_info == model.sd_checkpoint_info and model: return base_vae -- cgit v1.2.3 From af27b716e53671c52308d4e101214b0fd4fd5e80 Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Fri, 11 Aug 2023 12:22:11 -0400 Subject: Fix color correction by converting image to RGB --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index c048ca25..131c4c3c 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -57,7 +57,7 @@ def apply_color_correction(correction, original_image): image = blendLayers(image, original_image, BlendType.LUMINOSITY) - return image + return image.convert('RGB') def apply_overlay(image, paste_loc, index, overlays): -- cgit v1.2.3 From 64311faa6848d641cc452115e4e1eb47d2a7b519 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 12 Aug 2023 12:39:59 +0300 Subject: put refiner into main UI, into the new accordions section add VAE from main model into infotext, not from refiner model option to make scripts UI without gr.Group fix inconsistencies with refiner when usings samplers that do more denoising than steps --- modules/processing.py | 22 ++++++++----- modules/processing_scripts/refiner.py | 55 +++++++++++++++++++++++++++++++++ modules/scripts.py | 24 ++++++++++----- modules/sd_models.py | 3 ++ modules/sd_samplers_cfg_denoiser.py | 6 +++- modules/sd_samplers_common.py | 40 ++++++++++++++---------- modules/sd_samplers_kdiffusion.py | 3 +- modules/shared_items.py | 4 +-- modules/shared_options.py | 2 -- modules/ui.py | 58 ++++++++++++++++++++--------------- modules/ui_components.py | 18 ++++++++--- style.css | 32 +++++++++++-------- 12 files changed, 188 insertions(+), 79 deletions(-) create mode 100644 modules/processing_scripts/refiner.py (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 131c4c3c..5996cbac 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -373,9 +373,10 @@ class StableDiffusionProcessing: negative_prompts = prompt_parser.SdConditioning(self.negative_prompts, width=self.width, height=self.height, is_negative_prompt=True) sampler_config = sd_samplers.find_sampler_config(self.sampler_name) - self.step_multiplier = 2 if sampler_config and sampler_config.options.get("second_order", False) else 1 - self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, self.steps * self.step_multiplier, [self.cached_uc], self.extra_network_data) - self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, self.steps * self.step_multiplier, [self.cached_c], self.extra_network_data) + total_steps = sampler_config.total_steps(self.steps) if sampler_config else self.steps + self.step_multiplier = total_steps // self.steps + self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, total_steps, [self.cached_uc], self.extra_network_data) + self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, total_steps, [self.cached_c], self.extra_network_data) def get_conds(self): return self.c, self.uc @@ -579,8 +580,8 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Size": f"{p.width}x{p.height}", "Model hash": getattr(p, 'sd_model_hash', None if not opts.add_model_hash_to_info or not shared.sd_model.sd_model_hash else shared.sd_model.sd_model_hash), "Model": (None if not opts.add_model_name_to_info else shared.sd_model.sd_checkpoint_info.name_for_extra), - "VAE hash": sd_vae.get_loaded_vae_hash() if opts.add_model_hash_to_info else None, - "VAE": sd_vae.get_loaded_vae_name() if opts.add_model_name_to_info else None, + "VAE hash": p.loaded_vae_hash if opts.add_model_hash_to_info else None, + "VAE": p.loaded_vae_name if opts.add_model_name_to_info else None, "Variation seed": (None if p.subseed_strength == 0 else (p.all_subseeds[0] if use_main_prompt else all_subseeds[index])), "Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength), "Seed resize from": (None if p.seed_resize_from_w <= 0 or p.seed_resize_from_h <= 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"), @@ -669,6 +670,9 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.tiling is None: p.tiling = opts.tiling + p.loaded_vae_name = sd_vae.get_loaded_vae_name() + p.loaded_vae_hash = sd_vae.get_loaded_vae_hash() + modules.sd_hijack.model_hijack.apply_circular(p.tiling) modules.sd_hijack.model_hijack.clear_comments() @@ -1188,8 +1192,12 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): hr_prompts = prompt_parser.SdConditioning(self.hr_prompts, width=self.hr_upscale_to_x, height=self.hr_upscale_to_y) hr_negative_prompts = prompt_parser.SdConditioning(self.hr_negative_prompts, width=self.hr_upscale_to_x, height=self.hr_upscale_to_y, is_negative_prompt=True) - self.hr_uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, hr_negative_prompts, self.steps * self.step_multiplier, [self.cached_hr_uc, self.cached_uc], self.hr_extra_network_data) - self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, hr_prompts, self.steps * self.step_multiplier, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) + sampler_config = sd_samplers.find_sampler_config(self.hr_sampler_name or self.sampler_name) + steps = self.hr_second_pass_steps or self.steps + total_steps = sampler_config.total_steps(steps) if sampler_config else steps + + self.hr_uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, hr_negative_prompts, total_steps, [self.cached_hr_uc, self.cached_uc], self.hr_extra_network_data) + self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, hr_prompts, total_steps, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) def setup_conds(self): super().setup_conds() diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py new file mode 100644 index 00000000..5a82991a --- /dev/null +++ b/modules/processing_scripts/refiner.py @@ -0,0 +1,55 @@ +import gradio as gr + +from modules import scripts, sd_models +from modules.ui_common import create_refresh_button +from modules.ui_components import InputAccordion + + +class ScriptRefiner(scripts.Script): + section = "accordions" + create_group = False + + def __init__(self): + pass + + def title(self): + return "Refiner" + + def show(self, is_img2img): + return scripts.AlwaysVisible + + def ui(self, is_img2img): + with InputAccordion(False, label="Refiner", elem_id=self.elem_id("enable")) as enable_refiner: + with gr.Row(): + refiner_checkpoint = gr.Dropdown(label='Checkpoint', elem_id=self.elem_id("checkpoint"), choices=sd_models.checkpoint_tiles(), value='', tooltip="switch to another model in the middle of generation") + create_refresh_button(refiner_checkpoint, sd_models.list_models, lambda: {"choices": sd_models.checkpoint_tiles()}, self.elem_id("checkpoint_refresh")) + + refiner_switch_at = gr.Slider(value=0.8, label="Switch at", minimum=0.01, maximum=1.0, step=0.01, elem_id=self.elem_id("switch_at"), tooltip="fraction of sampling steps when the swtch to refiner model should happen; 1=never, 0.5=switch in the middle of generation") + + def lookup_checkpoint(title): + info = sd_models.get_closet_checkpoint_match(title) + return None if info is None else info.title + + self.infotext_fields = [ + (enable_refiner, lambda d: 'Refiner' in d), + (refiner_checkpoint, lambda d: lookup_checkpoint(d.get('Refiner'))), + (refiner_switch_at, 'Refiner switch at'), + ] + + return enable_refiner, refiner_checkpoint, refiner_switch_at + + def before_process(self, p, enable_refiner, refiner_checkpoint, refiner_switch_at): + # the actual implementation is in sd_samplers_common.py, apply_refiner + + p.refiner_checkpoint_info = None + p.refiner_switch_at = None + + if not enable_refiner or refiner_checkpoint in (None, "", "None"): + return + + refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(refiner_checkpoint) + if refiner_checkpoint_info is None: + raise Exception(f'Could not find checkpoint with name {refiner_checkpoint}') + + p.refiner_checkpoint_info = refiner_checkpoint_info + p.refiner_switch_at = refiner_switch_at diff --git a/modules/scripts.py b/modules/scripts.py index f7d060aa..51da732a 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -37,7 +37,10 @@ class Script: is_img2img = False group = None - """A gr.Group component that has all script's UI inside it""" + """A gr.Group component that has all script's UI inside it.""" + + create_group = True + """If False, for alwayson scripts, a group component will not be created.""" infotext_fields = None """if set in ui(), this is a list of pairs of gradio component + text; the text will be used when @@ -232,6 +235,7 @@ class Script: """ pass + current_basedir = paths.script_path @@ -250,7 +254,7 @@ postprocessing_scripts_data = [] ScriptClassData = namedtuple("ScriptClassData", ["script_class", "path", "basedir", "module"]) -def list_scripts(scriptdirname, extension): +def list_scripts(scriptdirname, extension, *, include_extensions=True): scripts_list = [] basedir = os.path.join(paths.script_path, scriptdirname) @@ -258,8 +262,9 @@ def list_scripts(scriptdirname, extension): for filename in sorted(os.listdir(basedir)): scripts_list.append(ScriptFile(paths.script_path, filename, os.path.join(basedir, filename))) - for ext in extensions.active(): - scripts_list += ext.list_files(scriptdirname, extension) + if include_extensions: + for ext in extensions.active(): + scripts_list += ext.list_files(scriptdirname, extension) scripts_list = [x for x in scripts_list if os.path.splitext(x.path)[1].lower() == extension and os.path.isfile(x.path)] @@ -288,7 +293,7 @@ def load_scripts(): postprocessing_scripts_data.clear() script_callbacks.clear_callbacks() - scripts_list = list_scripts("scripts", ".py") + scripts_list = list_scripts("scripts", ".py") + list_scripts("modules/processing_scripts", ".py", include_extensions=False) syspath = sys.path @@ -429,10 +434,13 @@ class ScriptRunner: if script.alwayson and script.section != section: continue - with gr.Group(visible=script.alwayson) as group: - self.create_script_ui(script) + if script.create_group: + with gr.Group(visible=script.alwayson) as group: + self.create_script_ui(script) - script.group = group + script.group = group + else: + self.create_script_ui(script) def prepare_ui(self): self.inputs = [None] diff --git a/modules/sd_models.py b/modules/sd_models.py index a178adca..f6fbdcd6 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -147,6 +147,9 @@ re_strip_checksum = re.compile(r"\s*\[[^]]+]\s*$") def get_closet_checkpoint_match(search_string): + if not search_string: + return None + checkpoint_info = checkpoint_aliases.get(search_string, None) if checkpoint_info is not None: return checkpoint_info diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index a532e013..113425b2 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -45,6 +45,11 @@ class CFGDenoiser(torch.nn.Module): self.nmask = None self.init_latent = None self.steps = None + """number of steps as specified by user in UI""" + + self.total_steps = None + """expected number of calls to denoiser calculated from self.steps and specifics of the selected sampler""" + self.step = 0 self.image_cfg_scale = None self.padded_cond_uncond = False @@ -56,7 +61,6 @@ class CFGDenoiser(torch.nn.Module): def inner_model(self): raise NotImplementedError() - def combine_denoised(self, x_out, conds_list, uncond, cond_scale): denoised_uncond = x_out[-uncond.shape[0]:] denoised = torch.clone(denoised_uncond) diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 35c4d657..85f3c7e0 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -7,7 +7,16 @@ from modules import devices, images, sd_vae_approx, sd_samplers, sd_vae_taesd, s from modules.shared import opts, state import k_diffusion.sampling -SamplerData = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options']) + +SamplerDataTuple = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options']) + + +class SamplerData(SamplerDataTuple): + def total_steps(self, steps): + if self.options.get("second_order", False): + steps = steps * 2 + + return steps def setup_img2img_steps(p, steps=None): @@ -131,31 +140,26 @@ def replace_torchsde_browinan(): replace_torchsde_browinan() -def apply_refiner(sampler): - completed_ratio = sampler.step / sampler.steps +def apply_refiner(cfg_denoiser): + completed_ratio = cfg_denoiser.step / cfg_denoiser.total_steps + refiner_switch_at = cfg_denoiser.p.refiner_switch_at + refiner_checkpoint_info = cfg_denoiser.p.refiner_checkpoint_info - if completed_ratio <= shared.opts.sd_refiner_switch_at: + if refiner_switch_at is not None and completed_ratio <= refiner_switch_at: return False - if shared.opts.sd_refiner_checkpoint == "None": + if refiner_checkpoint_info is None or shared.sd_model.sd_checkpoint_info == refiner_checkpoint_info: return False - if shared.sd_model.sd_checkpoint_info.title == shared.opts.sd_refiner_checkpoint: - return False - - refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(shared.opts.sd_refiner_checkpoint) - if refiner_checkpoint_info is None: - raise Exception(f'Could not find checkpoint with name {shared.opts.sd_refiner_checkpoint}') - - sampler.p.extra_generation_params['Refiner'] = refiner_checkpoint_info.short_title - sampler.p.extra_generation_params['Refiner switch at'] = shared.opts.sd_refiner_switch_at + cfg_denoiser.p.extra_generation_params['Refiner'] = refiner_checkpoint_info.short_title + cfg_denoiser.p.extra_generation_params['Refiner switch at'] = refiner_switch_at with sd_models.SkipWritingToConfig(): sd_models.reload_model_weights(info=refiner_checkpoint_info) devices.torch_gc() - sampler.p.setup_conds() - sampler.update_inner_model() + cfg_denoiser.p.setup_conds() + cfg_denoiser.update_inner_model() return True @@ -192,7 +196,7 @@ class Sampler: self.sampler_noises = None self.stop_at = None self.eta = None - self.config = None # set by the function calling the constructor + self.config: SamplerData = None # set by the function calling the constructor self.last_latent = None self.s_min_uncond = None self.s_churn = 0.0 @@ -208,6 +212,7 @@ class Sampler: self.p = None self.model_wrap_cfg = None self.sampler_extra_args = None + self.options = {} def callback_state(self, d): step = d['i'] @@ -220,6 +225,7 @@ class Sampler: def launch_sampling(self, steps, func): self.model_wrap_cfg.steps = steps + self.model_wrap_cfg.total_steps = self.config.total_steps(steps) state.sampling_steps = steps state.sampling_step = 0 diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index d10fe12e..1f8e9c4b 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -64,9 +64,10 @@ class CFGDenoiserKDiffusion(sd_samplers_cfg_denoiser.CFGDenoiser): class KDiffusionSampler(sd_samplers_common.Sampler): - def __init__(self, funcname, sd_model): + def __init__(self, funcname, sd_model, options=None): super().__init__(funcname) + self.options = options or {} self.func = funcname if callable(funcname) else getattr(k_diffusion.sampling, self.funcname) self.model_wrap_cfg = CFGDenoiserKDiffusion(self) diff --git a/modules/shared_items.py b/modules/shared_items.py index e4ec40a8..754166d2 100644 --- a/modules/shared_items.py +++ b/modules/shared_items.py @@ -69,8 +69,8 @@ def reload_hypernetworks(): ui_reorder_categories_builtin_items = [ "inpaint", "sampler", + "accordions", "checkboxes", - "hires_fix", "dimensions", "cfg", "seed", @@ -86,7 +86,7 @@ def ui_reorder_categories(): sections = {} for script in scripts.scripts_txt2img.scripts + scripts.scripts_img2img.scripts: - if isinstance(script.section, str): + if isinstance(script.section, str) and script.section not in ui_reorder_categories_builtin_items: sections[script.section] = 1 yield from sections diff --git a/modules/shared_options.py b/modules/shared_options.py index 1e5b64ea..9ae51f18 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -140,8 +140,6 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "upcast_attn": OptionInfo(False, "Upcast cross attention layer to float32"), "randn_source": OptionInfo("GPU", "Random number generator source.", gr.Radio, {"choices": ["GPU", "CPU", "NV"]}).info("changes seeds drastically; use CPU to produce the same picture across different videocard vendors; use NV to produce same picture as on NVidia videocards"), "tiling": OptionInfo(False, "Tiling", infotext='Tiling').info("produce a tileable picture"), - "sd_refiner_checkpoint": OptionInfo("None", "Refiner checkpoint", gr.Dropdown, lambda: {"choices": ["None"] + shared_items.list_checkpoint_tiles()}, refresh=shared_items.refresh_checkpoints, infotext="Refiner").info("switch to another model in the middle of generation"), - "sd_refiner_switch_at": OptionInfo(1.0, "Refiner switch at", gr.Slider, {"minimum": 0.01, "maximum": 1.0, "step": 0.01}, infotext='Refiner switch at').info("fraction of sampling steps when the swtch to refiner model should happen; 1=never, 0.5=switch in the middle of generation"), })) options_templates.update(options_section(('sdxl', "Stable Diffusion XL"), { diff --git a/modules/ui.py b/modules/ui.py index 05292734..3321b94d 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -438,35 +438,38 @@ def create_ui(): with FormRow(elem_classes="checkboxes-row", variant="compact"): pass - elif category == "hires_fix": - with InputAccordion(False, label="Hires. fix") as enable_hr: - with enable_hr.extra(): - hr_final_resolution = FormHTML(value="", elem_id="txtimg_hr_finalres", label="Upscaled resolution", interactive=False, min_width=0) + elif category == "accordions": + with gr.Row(elem_id="txt2img_accordions", elem_classes="accordions"): + with InputAccordion(False, label="Hires. fix", elem_id="txt2img_hr") as enable_hr: + with enable_hr.extra(): + hr_final_resolution = FormHTML(value="", elem_id="txtimg_hr_finalres", label="Upscaled resolution", interactive=False, min_width=0) - with FormRow(elem_id="txt2img_hires_fix_row1", variant="compact"): - hr_upscaler = gr.Dropdown(label="Upscaler", elem_id="txt2img_hr_upscaler", choices=[*shared.latent_upscale_modes, *[x.name for x in shared.sd_upscalers]], value=shared.latent_upscale_default_mode) - hr_second_pass_steps = gr.Slider(minimum=0, maximum=150, step=1, label='Hires steps', value=0, elem_id="txt2img_hires_steps") - denoising_strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label='Denoising strength', value=0.7, elem_id="txt2img_denoising_strength") + with FormRow(elem_id="txt2img_hires_fix_row1", variant="compact"): + hr_upscaler = gr.Dropdown(label="Upscaler", elem_id="txt2img_hr_upscaler", choices=[*shared.latent_upscale_modes, *[x.name for x in shared.sd_upscalers]], value=shared.latent_upscale_default_mode) + hr_second_pass_steps = gr.Slider(minimum=0, maximum=150, step=1, label='Hires steps', value=0, elem_id="txt2img_hires_steps") + denoising_strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label='Denoising strength', value=0.7, elem_id="txt2img_denoising_strength") - with FormRow(elem_id="txt2img_hires_fix_row2", variant="compact"): - hr_scale = gr.Slider(minimum=1.0, maximum=4.0, step=0.05, label="Upscale by", value=2.0, elem_id="txt2img_hr_scale") - hr_resize_x = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize width to", value=0, elem_id="txt2img_hr_resize_x") - hr_resize_y = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize height to", value=0, elem_id="txt2img_hr_resize_y") + with FormRow(elem_id="txt2img_hires_fix_row2", variant="compact"): + hr_scale = gr.Slider(minimum=1.0, maximum=4.0, step=0.05, label="Upscale by", value=2.0, elem_id="txt2img_hr_scale") + hr_resize_x = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize width to", value=0, elem_id="txt2img_hr_resize_x") + hr_resize_y = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize height to", value=0, elem_id="txt2img_hr_resize_y") - with FormRow(elem_id="txt2img_hires_fix_row3", variant="compact", visible=opts.hires_fix_show_sampler) as hr_sampler_container: + with FormRow(elem_id="txt2img_hires_fix_row3", variant="compact", visible=opts.hires_fix_show_sampler) as hr_sampler_container: - hr_checkpoint_name = gr.Dropdown(label='Hires checkpoint', elem_id="hr_checkpoint", choices=["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True), value="Use same checkpoint") - create_refresh_button(hr_checkpoint_name, modules.sd_models.list_models, lambda: {"choices": ["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True)}, "hr_checkpoint_refresh") + hr_checkpoint_name = gr.Dropdown(label='Hires checkpoint', elem_id="hr_checkpoint", choices=["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True), value="Use same checkpoint") + create_refresh_button(hr_checkpoint_name, modules.sd_models.list_models, lambda: {"choices": ["Use same checkpoint"] + modules.sd_models.checkpoint_tiles(use_short=True)}, "hr_checkpoint_refresh") - hr_sampler_name = gr.Dropdown(label='Hires sampling method', elem_id="hr_sampler", choices=["Use same sampler"] + sd_samplers.visible_sampler_names(), value="Use same sampler") + hr_sampler_name = gr.Dropdown(label='Hires sampling method', elem_id="hr_sampler", choices=["Use same sampler"] + sd_samplers.visible_sampler_names(), value="Use same sampler") - with FormRow(elem_id="txt2img_hires_fix_row4", variant="compact", visible=opts.hires_fix_show_prompts) as hr_prompts_container: - with gr.Column(scale=80): - with gr.Row(): - hr_prompt = gr.Textbox(label="Hires prompt", elem_id="hires_prompt", show_label=False, lines=3, placeholder="Prompt for hires fix pass.\nLeave empty to use the same prompt as in first pass.", elem_classes=["prompt"]) - with gr.Column(scale=80): - with gr.Row(): - hr_negative_prompt = gr.Textbox(label="Hires negative prompt", elem_id="hires_neg_prompt", show_label=False, lines=3, placeholder="Negative prompt for hires fix pass.\nLeave empty to use the same negative prompt as in first pass.", elem_classes=["prompt"]) + with FormRow(elem_id="txt2img_hires_fix_row4", variant="compact", visible=opts.hires_fix_show_prompts) as hr_prompts_container: + with gr.Column(scale=80): + with gr.Row(): + hr_prompt = gr.Textbox(label="Hires prompt", elem_id="hires_prompt", show_label=False, lines=3, placeholder="Prompt for hires fix pass.\nLeave empty to use the same prompt as in first pass.", elem_classes=["prompt"]) + with gr.Column(scale=80): + with gr.Row(): + hr_negative_prompt = gr.Textbox(label="Hires negative prompt", elem_id="hires_neg_prompt", show_label=False, lines=3, placeholder="Negative prompt for hires fix pass.\nLeave empty to use the same negative prompt as in first pass.", elem_classes=["prompt"]) + + scripts.scripts_txt2img.setup_ui_for_section(category) elif category == "batch": if not opts.dimensions_and_batch_together: @@ -482,7 +485,7 @@ def create_ui(): with FormGroup(elem_id="txt2img_script_container"): custom_inputs = scripts.scripts_txt2img.setup_ui() - else: + if category not in {"accordions"}: scripts.scripts_txt2img.setup_ui_for_section(category) hr_resolution_preview_inputs = [enable_hr, width, height, hr_scale, hr_resize_x, hr_resize_y] @@ -794,6 +797,10 @@ def create_ui(): with FormRow(elem_classes="checkboxes-row", variant="compact"): pass + elif category == "accordions": + with gr.Row(elem_id="img2img_accordions", elem_classes="accordions"): + scripts.scripts_img2img.setup_ui_for_section(category) + elif category == "batch": if not opts.dimensions_and_batch_together: with FormRow(elem_id="img2img_column_batch"): @@ -836,7 +843,8 @@ def create_ui(): inputs=[], outputs=[inpaint_controls, mask_alpha], ) - else: + + if category not in {"accordions"}: scripts.scripts_img2img.setup_ui_for_section(category) img2img_gallery, generation_info, html_info, html_log = create_output_panel("img2img", opts.outdir_img2img_samples) diff --git a/modules/ui_components.py b/modules/ui_components.py index bfe2fbd9..d08b2b99 100644 --- a/modules/ui_components.py +++ b/modules/ui_components.py @@ -87,13 +87,23 @@ class InputAccordion(gr.Checkbox): self.accordion_id = f"input-accordion-{InputAccordion.global_index}" InputAccordion.global_index += 1 - kwargs['elem_id'] = self.accordion_id + "-checkbox" - kwargs['visible'] = False - super().__init__(value, **kwargs) + kwargs_checkbox = { + **kwargs, + "elem_id": f"{self.accordion_id}-checkbox", + "visible": False, + } + super().__init__(value, **kwargs_checkbox) self.change(fn=None, _js='function(checked){ inputAccordionChecked("' + self.accordion_id + '", checked); }', inputs=[self]) - self.accordion = gr.Accordion(kwargs.get('label', 'Accordion'), open=value, elem_id=self.accordion_id, elem_classes=['input-accordion']) + kwargs_accordion = { + **kwargs, + "elem_id": self.accordion_id, + "label": kwargs.get('label', 'Accordion'), + "elem_classes": ['input-accordion'], + "open": value, + } + self.accordion = gr.Accordion(**kwargs_accordion) def extra(self): """Allows you to put something into the label of the accordion. diff --git a/style.css b/style.css index 4cdce87c..260b1056 100644 --- a/style.css +++ b/style.css @@ -166,16 +166,6 @@ a{ color: var(--button-secondary-text-color-hover); } -.checkboxes-row{ - margin-bottom: 0.5em; - margin-left: 0em; -} -.checkboxes-row > div{ - flex: 0; - white-space: nowrap; - min-width: auto !important; -} - button.custom-button{ border-radius: var(--button-large-radius); padding: var(--button-large-padding); @@ -352,7 +342,7 @@ div.block.gradio-accordion { } div.dimensions-tools{ - min-width: 0 !important; + min-width: 1.6em !important; max-width: fit-content; flex-direction: column; place-content: center; @@ -1012,10 +1002,28 @@ div.block.gradio-box.popup-dialog > div:last-child, .popup-dialog > div:last-chi } div.block.input-accordion{ - margin-bottom: 0.4em; + } .input-accordion-extra{ flex: 0 0 auto !important; margin: 0 0.5em 0 auto; } + +div.accordions > div.input-accordion{ + min-width: fit-content !important; +} + +div.accordions > div.gradio-accordion .label-wrap span{ + white-space: nowrap; + margin-right: 0.25em; +} + +div.accordions{ + gap: 0.5em; +} + +div.accordions > div.input-accordion.input-accordion-open{ + flex: 1 auto; +} + -- cgit v1.2.3 From b293ed30610c040e621e1840d63047ae298f0650 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 12 Aug 2023 12:54:32 +0300 Subject: make it possible to use hires fix together with refiner --- modules/processing.py | 6 ++++++ modules/processing_scripts/refiner.py | 2 +- modules/sd_samplers_common.py | 3 +++ 3 files changed, 10 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 5996cbac..6ad105d7 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -1200,6 +1200,12 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): self.hr_c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, hr_prompts, total_steps, [self.cached_hr_c, self.cached_c], self.hr_extra_network_data) def setup_conds(self): + if self.is_hr_pass: + # if we are in hr pass right now, the call is being made from the refiner, and we don't need to setup firstpass cons or switch model + self.hr_c = None + self.calculate_hr_conds() + return + super().setup_conds() self.hr_uc = None diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index 5a82991a..773ec5d0 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -24,7 +24,7 @@ class ScriptRefiner(scripts.Script): refiner_checkpoint = gr.Dropdown(label='Checkpoint', elem_id=self.elem_id("checkpoint"), choices=sd_models.checkpoint_tiles(), value='', tooltip="switch to another model in the middle of generation") create_refresh_button(refiner_checkpoint, sd_models.list_models, lambda: {"choices": sd_models.checkpoint_tiles()}, self.elem_id("checkpoint_refresh")) - refiner_switch_at = gr.Slider(value=0.8, label="Switch at", minimum=0.01, maximum=1.0, step=0.01, elem_id=self.elem_id("switch_at"), tooltip="fraction of sampling steps when the swtch to refiner model should happen; 1=never, 0.5=switch in the middle of generation") + refiner_switch_at = gr.Slider(value=0.8, label="Switch at", minimum=0.01, maximum=1.0, step=0.01, elem_id=self.elem_id("switch_at"), tooltip="fraction of sampling steps when the switch to refiner model should happen; 1=never, 0.5=switch in the middle of generation") def lookup_checkpoint(title): info = sd_models.get_closet_checkpoint_match(title) diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 85f3c7e0..40c7aae0 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -151,6 +151,9 @@ def apply_refiner(cfg_denoiser): if refiner_checkpoint_info is None or shared.sd_model.sd_checkpoint_info == refiner_checkpoint_info: return False + if getattr(cfg_denoiser.p, "enable_hr", False) and not cfg_denoiser.p.is_hr_pass: + return False + cfg_denoiser.p.extra_generation_params['Refiner'] = refiner_checkpoint_info.short_title cfg_denoiser.p.extra_generation_params['Refiner switch at'] = refiner_switch_at -- cgit v1.2.3 From fa9370b7411166c19e8e386400dc4e6082f47b2d Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 13 Aug 2023 06:07:30 +0300 Subject: add refiner to StableDiffusionProcessing class write out correct model name in infotext, rather than the refiner model --- modules/processing.py | 38 +++++++++++++++++++++++++++-------- modules/processing_scripts/refiner.py | 16 +++++---------- modules/sd_samplers_common.py | 2 +- 3 files changed, 36 insertions(+), 20 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 6ad105d7..b47ddaa8 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -111,7 +111,7 @@ class StableDiffusionProcessing: cached_uc = [None, None] cached_c = [None, None] - def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = None, tiling: bool = None, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = None, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): + def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = None, tiling: bool = None, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = None, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, refiner_checkpoint: str = None, refiner_switch_at: float = None, script_args: list = None): if sampler_index is not None: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) @@ -153,10 +153,14 @@ class StableDiffusionProcessing: self.s_noise = s_noise if s_noise is not None else opts.s_noise self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} self.override_settings_restore_afterwards = override_settings_restore_afterwards + self.refiner_checkpoint = refiner_checkpoint + self.refiner_switch_at = refiner_switch_at + self.is_using_inpainting_conditioning = False self.disable_extra_networks = False self.token_merging_ratio = 0 self.token_merging_ratio_hr = 0 + self.refiner_checkpoint_info = None if not seed_enable_extras: self.subseed = -1 @@ -191,6 +195,11 @@ class StableDiffusionProcessing: self.user = None + self.sd_model_name = None + self.sd_model_hash = None + self.sd_vae_name = None + self.sd_vae_hash = None + @property def sd_model(self): return shared.sd_model @@ -408,7 +417,10 @@ class Processed: self.batch_size = p.batch_size self.restore_faces = p.restore_faces self.face_restoration_model = opts.face_restoration_model if p.restore_faces else None - self.sd_model_hash = shared.sd_model.sd_model_hash + self.sd_model_name = p.sd_model_name + self.sd_model_hash = p.sd_model_hash + self.sd_vae_name = p.sd_vae_name + self.sd_vae_hash = p.sd_vae_hash self.seed_resize_from_w = p.seed_resize_from_w self.seed_resize_from_h = p.seed_resize_from_h self.denoising_strength = getattr(p, 'denoising_strength', None) @@ -459,7 +471,10 @@ class Processed: "batch_size": self.batch_size, "restore_faces": self.restore_faces, "face_restoration_model": self.face_restoration_model, + "sd_model_name": self.sd_model_name, "sd_model_hash": self.sd_model_hash, + "sd_vae_name": self.sd_vae_name, + "sd_vae_hash": self.sd_vae_hash, "seed_resize_from_w": self.seed_resize_from_w, "seed_resize_from_h": self.seed_resize_from_h, "denoising_strength": self.denoising_strength, @@ -578,10 +593,10 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Seed": p.all_seeds[0] if use_main_prompt else all_seeds[index], "Face restoration": opts.face_restoration_model if p.restore_faces else None, "Size": f"{p.width}x{p.height}", - "Model hash": getattr(p, 'sd_model_hash', None if not opts.add_model_hash_to_info or not shared.sd_model.sd_model_hash else shared.sd_model.sd_model_hash), - "Model": (None if not opts.add_model_name_to_info else shared.sd_model.sd_checkpoint_info.name_for_extra), - "VAE hash": p.loaded_vae_hash if opts.add_model_hash_to_info else None, - "VAE": p.loaded_vae_name if opts.add_model_name_to_info else None, + "Model hash": p.sd_model_hash if opts.add_model_hash_to_info else None, + "Model": p.sd_model_name if opts.add_model_name_to_info else None, + "VAE hash": p.sd_vae_hash if opts.add_model_hash_to_info else None, + "VAE": p.sd_vae_name if opts.add_model_name_to_info else None, "Variation seed": (None if p.subseed_strength == 0 else (p.all_subseeds[0] if use_main_prompt else all_subseeds[index])), "Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength), "Seed resize from": (None if p.seed_resize_from_w <= 0 or p.seed_resize_from_h <= 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"), @@ -670,8 +685,15 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.tiling is None: p.tiling = opts.tiling - p.loaded_vae_name = sd_vae.get_loaded_vae_name() - p.loaded_vae_hash = sd_vae.get_loaded_vae_hash() + if p.refiner_checkpoint not in (None, "", "None"): + p.refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(p.refiner_checkpoint) + if p.refiner_checkpoint_info is None: + raise Exception(f'Could not find checkpoint with name {p.refiner_checkpoint}') + + p.sd_model_name = shared.sd_model.sd_checkpoint_info.name_for_extra + p.sd_model_hash = shared.sd_model.sd_model_hash + p.sd_vae_name = sd_vae.get_loaded_vae_name() + p.sd_vae_hash = sd_vae.get_loaded_vae_hash() modules.sd_hijack.model_hijack.apply_circular(p.tiling) modules.sd_hijack.model_hijack.clear_comments() diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index 773ec5d0..7b946d05 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -41,15 +41,9 @@ class ScriptRefiner(scripts.Script): def before_process(self, p, enable_refiner, refiner_checkpoint, refiner_switch_at): # the actual implementation is in sd_samplers_common.py, apply_refiner - p.refiner_checkpoint_info = None - p.refiner_switch_at = None - if not enable_refiner or refiner_checkpoint in (None, "", "None"): - return - - refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(refiner_checkpoint) - if refiner_checkpoint_info is None: - raise Exception(f'Could not find checkpoint with name {refiner_checkpoint}') - - p.refiner_checkpoint_info = refiner_checkpoint_info - p.refiner_switch_at = refiner_switch_at + p.refiner_checkpoint_info = None + p.refiner_switch_at = None + else: + p.refiner_checkpoint = refiner_checkpoint + p.refiner_switch_at = refiner_switch_at diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 40c7aae0..380cdd5f 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -145,7 +145,7 @@ def apply_refiner(cfg_denoiser): refiner_switch_at = cfg_denoiser.p.refiner_switch_at refiner_checkpoint_info = cfg_denoiser.p.refiner_checkpoint_info - if refiner_switch_at is not None and completed_ratio <= refiner_switch_at: + if refiner_switch_at is not None and completed_ratio < refiner_switch_at: return False if refiner_checkpoint_info is None or shared.sd_model.sd_checkpoint_info == refiner_checkpoint_info: -- cgit v1.2.3 From 599f61a1e0bddf463dd3c6adb84509b3d9db1941 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 13 Aug 2023 08:24:16 +0300 Subject: use dataclass for StableDiffusionProcessing --- modules/processing.py | 318 +++++++++++++++++++++++------------------- modules/sd_samplers_common.py | 5 +- 2 files changed, 176 insertions(+), 147 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b47ddaa8..007a4e05 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -1,9 +1,11 @@ +from __future__ import annotations import json import logging import math import os import sys import hashlib +from dataclasses import dataclass, field import torch import numpy as np @@ -11,7 +13,7 @@ from PIL import Image, ImageOps import random import cv2 from skimage import exposure -from typing import Any, Dict, List +from typing import Any import modules.sd_hijack from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng @@ -104,106 +106,126 @@ def txt2img_image_conditioning(sd_model, x, width, height): return x.new_zeros(x.shape[0], 5, 1, 1, dtype=x.dtype, device=x.device) +@dataclass(repr=False) class StableDiffusionProcessing: - """ - The first set of paramaters: sd_models -> do_not_reload_embeddings represent the minimum required to create a StableDiffusionProcessing - """ + sd_model: object = None + outpath_samples: str = None + outpath_grids: str = None + prompt: str = "" + prompt_for_display: str = None + negative_prompt: str = "" + styles: list[str] = field(default_factory=list) + seed: int = -1 + subseed: int = -1 + subseed_strength: float = 0 + seed_resize_from_h: int = -1 + seed_resize_from_w: int = -1 + seed_enable_extras: bool = True + sampler_name: str = None + batch_size: int = 1 + n_iter: int = 1 + steps: int = 50 + cfg_scale: float = 7.0 + width: int = 512 + height: int = 512 + restore_faces: bool = None + tiling: bool = None + do_not_save_samples: bool = False + do_not_save_grid: bool = False + extra_generation_params: dict[str, Any] = None + overlay_images: list = None + eta: float = None + do_not_reload_embeddings: bool = False + denoising_strength: float = 0 + ddim_discretize: str = None + s_min_uncond: float = None + s_churn: float = None + s_tmax: float = None + s_tmin: float = None + s_noise: float = None + override_settings: dict[str, Any] = None + override_settings_restore_afterwards: bool = True + sampler_index: int = None + refiner_checkpoint: str = None + refiner_switch_at: float = None + token_merging_ratio = 0 + token_merging_ratio_hr = 0 + disable_extra_networks: bool = False + + script_args: list = None + cached_uc = [None, None] cached_c = [None, None] - def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = None, tiling: bool = None, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_min_uncond: float = 0.0, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = None, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, refiner_checkpoint: str = None, refiner_switch_at: float = None, script_args: list = None): - if sampler_index is not None: + sampler: sd_samplers_common.Sampler | None = field(default=None, init=False) + is_using_inpainting_conditioning: bool = field(default=False, init=False) + paste_to: tuple | None = field(default=None, init=False) + + is_hr_pass: bool = field(default=False, init=False) + + c: tuple = field(default=None, init=False) + uc: tuple = field(default=None, init=False) + + rng: rng.ImageRNG | None = field(default=None, init=False) + step_multiplier: int = field(default=1, init=False) + color_corrections: list = field(default=None, init=False) + + scripts: list = field(default=None, init=False) + all_prompts: list = field(default=None, init=False) + all_negative_prompts: list = field(default=None, init=False) + all_seeds: list = field(default=None, init=False) + all_subseeds: list = field(default=None, init=False) + iteration: int = field(default=0, init=False) + main_prompt: str = field(default=None, init=False) + main_negative_prompt: str = field(default=None, init=False) + + prompts: list = field(default=None, init=False) + negative_prompts: list = field(default=None, init=False) + seeds: list = field(default=None, init=False) + subseeds: list = field(default=None, init=False) + extra_network_data: dict = field(default=None, init=False) + + user: str = field(default=None, init=False) + + sd_model_name: str = field(default=None, init=False) + sd_model_hash: str = field(default=None, init=False) + sd_vae_name: str = field(default=None, init=False) + sd_vae_hash: str = field(default=None, init=False) + + def __post_init__(self): + if self.sampler_index is not None: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) - self.outpath_samples: str = outpath_samples - self.outpath_grids: str = outpath_grids - self.prompt: str = prompt - self.prompt_for_display: str = None - self.negative_prompt: str = (negative_prompt or "") - self.styles: list = styles or [] - self.seed: int = seed - self.subseed: int = subseed - self.subseed_strength: float = subseed_strength - self.seed_resize_from_h: int = seed_resize_from_h - self.seed_resize_from_w: int = seed_resize_from_w - self.sampler_name: str = sampler_name - self.batch_size: int = batch_size - self.n_iter: int = n_iter - self.steps: int = steps - self.cfg_scale: float = cfg_scale - self.width: int = width - self.height: int = height - self.restore_faces: bool = restore_faces - self.tiling: bool = tiling - self.do_not_save_samples: bool = do_not_save_samples - self.do_not_save_grid: bool = do_not_save_grid - self.extra_generation_params: dict = extra_generation_params or {} - self.overlay_images = overlay_images - self.eta = eta - self.do_not_reload_embeddings = do_not_reload_embeddings - self.paste_to = None - self.color_corrections = None - self.denoising_strength: float = denoising_strength self.sampler_noise_scheduler_override = None - self.ddim_discretize = ddim_discretize or opts.ddim_discretize - self.s_min_uncond = s_min_uncond or opts.s_min_uncond - self.s_churn = s_churn or opts.s_churn - self.s_tmin = s_tmin or opts.s_tmin - self.s_tmax = (s_tmax if s_tmax is not None else opts.s_tmax) or float('inf') - self.s_noise = s_noise if s_noise is not None else opts.s_noise - self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts} - self.override_settings_restore_afterwards = override_settings_restore_afterwards - self.refiner_checkpoint = refiner_checkpoint - self.refiner_switch_at = refiner_switch_at - - self.is_using_inpainting_conditioning = False - self.disable_extra_networks = False - self.token_merging_ratio = 0 - self.token_merging_ratio_hr = 0 + self.s_min_uncond = self.s_min_uncond if self.s_min_uncond is not None else opts.s_min_uncond + self.s_churn = self.s_churn if self.s_churn is not None else opts.s_churn + self.s_tmin = self.s_tmin if self.s_tmin is not None else opts.s_tmin + self.s_tmax = (self.s_tmax if self.s_tmax is not None else opts.s_tmax) or float('inf') + self.s_noise = self.s_noise if self.s_noise is not None else opts.s_noise + + self.extra_generation_params = self.extra_generation_params or {} + self.override_settings = self.override_settings or {} + self.script_args = self.script_args or {} + self.refiner_checkpoint_info = None - if not seed_enable_extras: + if not self.seed_enable_extras: self.subseed = -1 self.subseed_strength = 0 self.seed_resize_from_h = 0 self.seed_resize_from_w = 0 - self.scripts = None - self.script_args = script_args - self.all_prompts = None - self.all_negative_prompts = None - self.all_seeds = None - self.all_subseeds = None - self.iteration = 0 - self.is_hr_pass = False - self.sampler = None - self.main_prompt = None - self.main_negative_prompt = None - - self.prompts = None - self.negative_prompts = None - self.extra_network_data = None - self.seeds = None - self.subseeds = None - - self.step_multiplier = 1 self.cached_uc = StableDiffusionProcessing.cached_uc self.cached_c = StableDiffusionProcessing.cached_c - self.uc = None - self.c = None - self.rng: rng.ImageRNG = None - - self.user = None - - self.sd_model_name = None - self.sd_model_hash = None - self.sd_vae_name = None - self.sd_vae_hash = None @property def sd_model(self): return shared.sd_model + @sd_model.setter + def sd_model(self, value): + pass + def txt2img_image_conditioning(self, x, width=None, height=None): self.is_using_inpainting_conditioning = self.sd_model.model.conditioning_key in {'hybrid', 'concat'} @@ -932,49 +954,51 @@ def old_hires_fix_first_pass_dimensions(width, height): return width, height +@dataclass(repr=False) class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): - sampler = None + enable_hr: bool = False + denoising_strength: float = 0.75 + firstphase_width: int = 0 + firstphase_height: int = 0 + hr_scale: float = 2.0 + hr_upscaler: str = None + hr_second_pass_steps: int = 0 + hr_resize_x: int = 0 + hr_resize_y: int = 0 + hr_checkpoint_name: str = None + hr_sampler_name: str = None + hr_prompt: str = '' + hr_negative_prompt: str = '' + cached_hr_uc = [None, None] cached_hr_c = [None, None] - def __init__(self, enable_hr: bool = False, denoising_strength: float = 0.75, firstphase_width: int = 0, firstphase_height: int = 0, hr_scale: float = 2.0, hr_upscaler: str = None, hr_second_pass_steps: int = 0, hr_resize_x: int = 0, hr_resize_y: int = 0, hr_checkpoint_name: str = None, hr_sampler_name: str = None, hr_prompt: str = '', hr_negative_prompt: str = '', **kwargs): - super().__init__(**kwargs) - self.enable_hr = enable_hr - self.denoising_strength = denoising_strength - self.hr_scale = hr_scale - self.hr_upscaler = hr_upscaler - self.hr_second_pass_steps = hr_second_pass_steps - self.hr_resize_x = hr_resize_x - self.hr_resize_y = hr_resize_y - self.hr_upscale_to_x = hr_resize_x - self.hr_upscale_to_y = hr_resize_y - self.hr_checkpoint_name = hr_checkpoint_name - self.hr_checkpoint_info = None - self.hr_sampler_name = hr_sampler_name - self.hr_prompt = hr_prompt - self.hr_negative_prompt = hr_negative_prompt - self.all_hr_prompts = None - self.all_hr_negative_prompts = None - self.latent_scale_mode = None - - if firstphase_width != 0 or firstphase_height != 0: + hr_checkpoint_info: dict = field(default=None, init=False) + hr_upscale_to_x: int = field(default=0, init=False) + hr_upscale_to_y: int = field(default=0, init=False) + truncate_x: int = field(default=0, init=False) + truncate_y: int = field(default=0, init=False) + applied_old_hires_behavior_to: tuple = field(default=None, init=False) + latent_scale_mode: dict = field(default=None, init=False) + hr_c: tuple | None = field(default=None, init=False) + hr_uc: tuple | None = field(default=None, init=False) + all_hr_prompts: list = field(default=None, init=False) + all_hr_negative_prompts: list = field(default=None, init=False) + hr_prompts: list = field(default=None, init=False) + hr_negative_prompts: list = field(default=None, init=False) + hr_extra_network_data: list = field(default=None, init=False) + + def __post_init__(self): + super().__post_init__() + + if self.firstphase_width != 0 or self.firstphase_height != 0: self.hr_upscale_to_x = self.width self.hr_upscale_to_y = self.height - self.width = firstphase_width - self.height = firstphase_height - - self.truncate_x = 0 - self.truncate_y = 0 - self.applied_old_hires_behavior_to = None - - self.hr_prompts = None - self.hr_negative_prompts = None - self.hr_extra_network_data = None + self.width = self.firstphase_width + self.height = self.firstphase_height self.cached_hr_uc = StableDiffusionProcessingTxt2Img.cached_hr_uc self.cached_hr_c = StableDiffusionProcessingTxt2Img.cached_hr_c - self.hr_c = None - self.hr_uc = None def calculate_target_resolution(self): if opts.use_old_hires_fix_width_height and self.applied_old_hires_behavior_to != (self.width, self.height): @@ -1252,7 +1276,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): return super().get_conds() - def parse_extra_network_prompts(self): res = super().parse_extra_network_prompts() @@ -1265,32 +1288,37 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): return res +@dataclass(repr=False) class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): - sampler = None - - def __init__(self, init_images: list = None, resize_mode: int = 0, denoising_strength: float = 0.75, image_cfg_scale: float = None, mask: Any = None, mask_blur: int = None, mask_blur_x: int = 4, mask_blur_y: int = 4, inpainting_fill: int = 0, inpaint_full_res: bool = True, inpaint_full_res_padding: int = 0, inpainting_mask_invert: int = 0, initial_noise_multiplier: float = None, **kwargs): - super().__init__(**kwargs) - - self.init_images = init_images - self.resize_mode: int = resize_mode - self.denoising_strength: float = denoising_strength - self.image_cfg_scale: float = image_cfg_scale if shared.sd_model.cond_stage_key == "edit" else None - self.init_latent = None - self.image_mask = mask - self.latent_mask = None - self.mask_for_overlay = None - self.mask_blur_x = mask_blur_x - self.mask_blur_y = mask_blur_y - if mask_blur is not None: - self.mask_blur = mask_blur - self.inpainting_fill = inpainting_fill - self.inpaint_full_res = inpaint_full_res - self.inpaint_full_res_padding = inpaint_full_res_padding - self.inpainting_mask_invert = inpainting_mask_invert - self.initial_noise_multiplier = opts.initial_noise_multiplier if initial_noise_multiplier is None else initial_noise_multiplier + init_images: list = None + resize_mode: int = 0 + denoising_strength: float = 0.75 + image_cfg_scale: float = None + mask: Any = None + mask_blur_x: int = 4 + mask_blur_y: int = 4 + mask_blur: int = None + inpainting_fill: int = 0 + inpaint_full_res: bool = True + inpaint_full_res_padding: int = 0 + inpainting_mask_invert: int = 0 + initial_noise_multiplier: float = None + latent_mask: Image = None + + image_mask: Any = field(default=None, init=False) + + nmask: torch.Tensor = field(default=None, init=False) + image_conditioning: torch.Tensor = field(default=None, init=False) + init_img_hash: str = field(default=None, init=False) + mask_for_overlay: Image = field(default=None, init=False) + init_latent: torch.Tensor = field(default=None, init=False) + + def __post_init__(self): + super().__post_init__() + + self.image_mask = self.mask self.mask = None - self.nmask = None - self.image_conditioning = None + self.initial_noise_multiplier = opts.initial_noise_multiplier if self.initial_noise_multiplier is None else self.initial_noise_multiplier @property def mask_blur(self): @@ -1300,15 +1328,13 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): @mask_blur.setter def mask_blur(self, value): - self.mask_blur_x = value - self.mask_blur_y = value - - @mask_blur.deleter - def mask_blur(self): - del self.mask_blur_x - del self.mask_blur_y + if isinstance(value, int): + self.mask_blur_x = value + self.mask_blur_y = value def init(self, all_prompts, all_seeds, all_subseeds): + self.image_cfg_scale: float = self.image_cfg_scale if shared.sd_model.cond_stage_key == "edit" else None + self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) crop_region = None diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 380cdd5f..09d1e11e 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -305,5 +305,8 @@ class Sampler: current_iter_seeds = p.all_seeds[p.iteration * p.batch_size:(p.iteration + 1) * p.batch_size] return BrownianTreeNoiseSampler(x, sigma_min, sigma_max, seed=current_iter_seeds) + def sample(self, p, x, conditioning, unconditional_conditioning, steps=None, image_conditioning=None): + raise NotImplementedError() - + def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None): + raise NotImplementedError() -- cgit v1.2.3 From 69f49c8d394220331eaa6609825b477eed60e0f4 Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Sun, 13 Aug 2023 04:40:34 -0400 Subject: Clear sampler before decoding images More significant VRAM reduction. --- modules/processing.py | 3 +++ 1 file changed, 3 insertions(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 007a4e05..16104f92 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -1192,6 +1192,9 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): sd_models.apply_token_merging(self.sd_model, self.get_token_merging_ratio()) + self.sampler = None + devices.torch_gc() + decoded_samples = decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True) self.is_hr_pass = False -- cgit v1.2.3 From d8419762c1454ba51baa710d9ce8e762efc056ef Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 13 Aug 2023 15:07:37 +0300 Subject: Lora: output warnings in UI rather than fail for unfitting loras; switch to logging for error output in console --- extensions-builtin/Lora/extra_networks_lora.py | 12 ++++- extensions-builtin/Lora/networks.py | 61 ++++++++++++++++---------- extensions-builtin/Lora/scripts/lora_script.py | 6 +-- modules/processing.py | 13 +++--- 4 files changed, 58 insertions(+), 34 deletions(-) (limited to 'modules/processing.py') diff --git a/extensions-builtin/Lora/extra_networks_lora.py b/extensions-builtin/Lora/extra_networks_lora.py index ba2945c6..32e32cab 100644 --- a/extensions-builtin/Lora/extra_networks_lora.py +++ b/extensions-builtin/Lora/extra_networks_lora.py @@ -1,4 +1,4 @@ -from modules import extra_networks, shared +from modules import extra_networks, shared, sd_hijack import networks @@ -6,9 +6,14 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): def __init__(self): super().__init__('lora') + self.errors = {} + """mapping of network names to the number of errors the network had during operation""" + def activate(self, p, params_list): additional = shared.opts.sd_lora + self.errors.clear() + if additional != "None" and additional in networks.available_networks and not any(x for x in params_list if x.items[0] == additional): p.all_prompts = [x + f"" for x in p.all_prompts] params_list.append(extra_networks.ExtraNetworkParams(items=[additional, shared.opts.extra_networks_default_multiplier])) @@ -56,4 +61,7 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): p.extra_generation_params["Lora hashes"] = ", ".join(network_hashes) def deactivate(self, p): - pass + if self.errors: + p.comment("Networks with errors: " + ", ".join(f"{k} ({v})" for k, v in self.errors.items())) + + self.errors.clear() diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index ba621139..c252ed9e 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -1,3 +1,4 @@ +import logging import os import re @@ -194,7 +195,7 @@ def load_network(name, network_on_disk): net.modules[key] = net_module if keys_failed_to_match: - print(f"Failed to match keys when loading network {network_on_disk.filename}: {keys_failed_to_match}") + logging.debug(f"Network {network_on_disk.filename} didn't match keys: {keys_failed_to_match}") return net @@ -207,7 +208,6 @@ def purge_networks_from_memory(): devices.torch_gc() - def load_networks(names, te_multipliers=None, unet_multipliers=None, dyn_dims=None): already_loaded = {} @@ -248,7 +248,7 @@ def load_networks(names, te_multipliers=None, unet_multipliers=None, dyn_dims=No if net is None: failed_to_load_networks.append(name) - print(f"Couldn't find network with name {name}") + logging.info(f"Couldn't find network with name {name}") continue net.te_multiplier = te_multipliers[i] if te_multipliers else 1.0 @@ -257,7 +257,7 @@ def load_networks(names, te_multipliers=None, unet_multipliers=None, dyn_dims=No loaded_networks.append(net) if failed_to_load_networks: - sd_hijack.model_hijack.comments.append("Failed to find networks: " + ", ".join(failed_to_load_networks)) + sd_hijack.model_hijack.comments.append("Networks not found: " + ", ".join(failed_to_load_networks)) purge_networks_from_memory() @@ -314,17 +314,22 @@ def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn for net in loaded_networks: module = net.modules.get(network_layer_name, None) if module is not None and hasattr(self, 'weight'): - with torch.no_grad(): - updown, ex_bias = module.calc_updown(self.weight) + try: + with torch.no_grad(): + updown, ex_bias = module.calc_updown(self.weight) - if len(self.weight.shape) == 4 and self.weight.shape[1] == 9: - # inpainting model. zero pad updown to make channel[1] 4 to 9 - updown = torch.nn.functional.pad(updown, (0, 0, 0, 0, 0, 5)) + if len(self.weight.shape) == 4 and self.weight.shape[1] == 9: + # inpainting model. zero pad updown to make channel[1] 4 to 9 + updown = torch.nn.functional.pad(updown, (0, 0, 0, 0, 0, 5)) - self.weight += updown - if ex_bias is not None and getattr(self, 'bias', None) is not None: - self.bias += ex_bias - continue + self.weight += updown + if ex_bias is not None and getattr(self, 'bias', None) is not None: + self.bias += ex_bias + except RuntimeError as e: + logging.debug(f"Network {net.name} layer {network_layer_name}: {e}") + extra_network_lora.errors[net.name] = extra_network_lora.errors.get(net.name, 0) + 1 + + continue module_q = net.modules.get(network_layer_name + "_q_proj", None) module_k = net.modules.get(network_layer_name + "_k_proj", None) @@ -332,21 +337,28 @@ def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn module_out = net.modules.get(network_layer_name + "_out_proj", None) if isinstance(self, torch.nn.MultiheadAttention) and module_q and module_k and module_v and module_out: - with torch.no_grad(): - updown_q = module_q.calc_updown(self.in_proj_weight) - updown_k = module_k.calc_updown(self.in_proj_weight) - updown_v = module_v.calc_updown(self.in_proj_weight) - updown_qkv = torch.vstack([updown_q, updown_k, updown_v]) - updown_out = module_out.calc_updown(self.out_proj.weight) - - self.in_proj_weight += updown_qkv - self.out_proj.weight += updown_out - continue + try: + with torch.no_grad(): + updown_q = module_q.calc_updown(self.in_proj_weight) + updown_k = module_k.calc_updown(self.in_proj_weight) + updown_v = module_v.calc_updown(self.in_proj_weight) + updown_qkv = torch.vstack([updown_q, updown_k, updown_v]) + updown_out = module_out.calc_updown(self.out_proj.weight) + + self.in_proj_weight += updown_qkv + self.out_proj.weight += updown_out + + except RuntimeError as e: + logging.debug(f"Network {net.name} layer {network_layer_name}: {e}") + extra_network_lora.errors[net.name] = extra_network_lora.errors.get(net.name, 0) + 1 + + continue if module is None: continue - print(f'failed to calculate network weights for layer {network_layer_name}') + logging.debug(f"Network {net.name} layer {network_layer_name}: couldn't find supported operation") + extra_network_lora.errors[net.name] = extra_network_lora.errors.get(net.name, 0) + 1 self.network_current_names = wanted_names @@ -519,6 +531,7 @@ def infotext_pasted(infotext, params): if added: params["Prompt"] += "\n" + "".join(added) +extra_network_lora = None available_networks = {} available_network_aliases = {} diff --git a/extensions-builtin/Lora/scripts/lora_script.py b/extensions-builtin/Lora/scripts/lora_script.py index dc307f8c..4c6e774a 100644 --- a/extensions-builtin/Lora/scripts/lora_script.py +++ b/extensions-builtin/Lora/scripts/lora_script.py @@ -23,9 +23,9 @@ def unload(): def before_ui(): ui_extra_networks.register_page(ui_extra_networks_lora.ExtraNetworksPageLora()) - extra_network = extra_networks_lora.ExtraNetworkLora() - extra_networks.register_extra_network(extra_network) - extra_networks.register_extra_network_alias(extra_network, "lyco") + networks.extra_network_lora = extra_networks_lora.ExtraNetworkLora() + extra_networks.register_extra_network(networks.extra_network_lora) + extra_networks.register_extra_network_alias(networks.extra_network_lora, "lyco") if not hasattr(torch.nn, 'Linear_forward_before_network'): diff --git a/modules/processing.py b/modules/processing.py index 007a4e05..10749aa2 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -157,6 +157,7 @@ class StableDiffusionProcessing: cached_uc = [None, None] cached_c = [None, None] + comments: dict = None sampler: sd_samplers_common.Sampler | None = field(default=None, init=False) is_using_inpainting_conditioning: bool = field(default=False, init=False) paste_to: tuple | None = field(default=None, init=False) @@ -196,6 +197,8 @@ class StableDiffusionProcessing: if self.sampler_index is not None: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) + self.comments = {} + self.sampler_noise_scheduler_override = None self.s_min_uncond = self.s_min_uncond if self.s_min_uncond is not None else opts.s_min_uncond self.s_churn = self.s_churn if self.s_churn is not None else opts.s_churn @@ -226,6 +229,9 @@ class StableDiffusionProcessing: def sd_model(self, value): pass + def comment(self, text): + self.comments[text] = 1 + def txt2img_image_conditioning(self, x, width=None, height=None): self.is_using_inpainting_conditioning = self.sd_model.model.conditioning_key in {'hybrid', 'concat'} @@ -429,7 +435,7 @@ class Processed: self.subseed = subseed self.subseed_strength = p.subseed_strength self.info = info - self.comments = comments + self.comments = "".join(f"{comment}\n" for comment in p.comments) self.width = p.width self.height = p.height self.sampler_name = p.sampler_name @@ -720,8 +726,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: modules.sd_hijack.model_hijack.apply_circular(p.tiling) modules.sd_hijack.model_hijack.clear_comments() - comments = {} - p.setup_prompts() if type(seed) == list: @@ -801,7 +805,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.setup_conds() for comment in model_hijack.comments: - comments[comment] = 1 + p.comment(comment) p.extra_generation_params.update(model_hijack.extra_generation_params) @@ -930,7 +934,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: images_list=output_images, seed=p.all_seeds[0], info=infotexts[0], - comments="".join(f"{comment}\n" for comment in comments), subseed=p.all_subseeds[0], index_of_first_image=index_of_first_image, infotexts=infotexts, -- cgit v1.2.3 From f093c9d39d0fe9951a8f5c570027cecc68778ef2 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 13 Aug 2023 17:31:10 +0300 Subject: fix broken XYZ plot seeds add new callback for scripts to be used before processing --- modules/processing.py | 32 ++++++++++++++++++++++++++++++-- modules/processing_scripts/seed.py | 2 +- modules/scripts.py | 17 ++++++++++++++++- 3 files changed, 47 insertions(+), 4 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index fdf49359..74366655 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -152,7 +152,9 @@ class StableDiffusionProcessing: token_merging_ratio_hr = 0 disable_extra_networks: bool = False - script_args: list = None + scripts_value: scripts.ScriptRunner = field(default=None, init=False) + script_args_value: list = field(default=None, init=False) + scripts_setup_complete: bool = field(default=False, init=False) cached_uc = [None, None] cached_c = [None, None] @@ -171,7 +173,6 @@ class StableDiffusionProcessing: step_multiplier: int = field(default=1, init=False) color_corrections: list = field(default=None, init=False) - scripts: list = field(default=None, init=False) all_prompts: list = field(default=None, init=False) all_negative_prompts: list = field(default=None, init=False) all_seeds: list = field(default=None, init=False) @@ -229,6 +230,33 @@ class StableDiffusionProcessing: def sd_model(self, value): pass + @property + def scripts(self): + return self.scripts_value + + @scripts.setter + def scripts(self, value): + self.scripts_value = value + + if self.scripts_value and self.script_args_value and not self.scripts_setup_complete: + self.setup_scripts() + + @property + def script_args(self): + return self.script_args_value + + @script_args.setter + def script_args(self, value): + self.script_args_value = value + + if self.scripts_value and self.script_args_value and not self.scripts_setup_complete: + self.setup_scripts() + + def setup_scripts(self): + self.scripts_setup_complete = True + + self.scripts.setup_scrips(self) + def comment(self, text): self.comments[text] = 1 diff --git a/modules/processing_scripts/seed.py b/modules/processing_scripts/seed.py index cc90775a..96b44dfb 100644 --- a/modules/processing_scripts/seed.py +++ b/modules/processing_scripts/seed.py @@ -58,7 +58,7 @@ class ScriptSeed(scripts.ScriptBuiltin): return self.seed, subseed, subseed_strength - def before_process(self, p, seed, subseed, subseed_strength): + def setup(self, p, seed, subseed, subseed_strength): p.seed = seed if subseed_strength > 0: diff --git a/modules/scripts.py b/modules/scripts.py index c6459b45..d4a9da94 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -106,9 +106,16 @@ class Script: pass + def setup(self, p, *args): + """For AlwaysVisible scripts, this function is called when the processing object is set up, before any processing starts. + args contains all values returned by components from ui(). + """ + pass + + def before_process(self, p, *args): """ - This function is called very early before processing begins for AlwaysVisible scripts. + This function is called very early during processing begins for AlwaysVisible scripts. You can modify the processing object (p) here, inject hooks, etc. args contains all values returned by components from ui() """ @@ -706,6 +713,14 @@ class ScriptRunner: except Exception: errors.report(f"Error running before_hr: {script.filename}", exc_info=True) + def setup_scrips(self, p): + for script in self.alwayson_scripts: + try: + script_args = p.script_args[script.args_from:script.args_to] + script.setup(p, *script_args) + except Exception: + errors.report(f"Error running setup: {script.filename}", exc_info=True) + scripts_txt2img: ScriptRunner = None scripts_img2img: ScriptRunner = None -- cgit v1.2.3 From c7c16f805c9ea0da42d1d993f2ea7bda48beba76 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 14 Aug 2023 09:48:40 +0300 Subject: repair /docs page --- modules/api/models.py | 9 +++++---- modules/processing.py | 3 ++- 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/api/models.py b/modules/api/models.py index 800c9b93..6a574771 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -50,10 +50,12 @@ class PydanticModelGenerator: additional_fields = None, ): def field_type_generator(k, v): - # field_type = str if not overrides.get(k) else overrides[k]["type"] - # print(k, v.annotation, v.default) field_type = v.annotation + if field_type == 'Image': + # images are sent as base64 strings via API + field_type = 'str' + return Optional[field_type] def merge_class_params(class_): @@ -63,7 +65,6 @@ class PydanticModelGenerator: parameters = {**parameters, **inspect.signature(classes.__init__).parameters} return parameters - self._model_name = model_name self._class_data = merge_class_params(class_instance) @@ -72,7 +73,7 @@ class PydanticModelGenerator: field=underscore(k), field_alias=k, field_type=field_type_generator(k, v), - field_value=v.default + field_value=None if isinstance(v.default, property) else v.default ) for (k,v) in self._class_data.items() if k not in API_NOT_ALLOWED ] diff --git a/modules/processing.py b/modules/processing.py index 74366655..69d365b8 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -114,7 +114,7 @@ class StableDiffusionProcessing: prompt: str = "" prompt_for_display: str = None negative_prompt: str = "" - styles: list[str] = field(default_factory=list) + styles: list[str] = None seed: int = -1 subseed: int = -1 subseed_strength: float = 0 @@ -199,6 +199,7 @@ class StableDiffusionProcessing: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) self.comments = {} + self.styles = [] self.sampler_noise_scheduler_override = None self.s_min_uncond = self.s_min_uncond if self.s_min_uncond is not None else opts.s_min_uncond -- cgit v1.2.3 From abbecb3e7363e422d6840fbb5746c74fd453ead5 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 14 Aug 2023 10:15:10 +0300 Subject: further repair the /docs page to not break styles with the attempted fix --- modules/processing.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 69d365b8..f34ba48a 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -199,7 +199,9 @@ class StableDiffusionProcessing: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) self.comments = {} - self.styles = [] + + if self.styles is None: + self.styles = [] self.sampler_noise_scheduler_override = None self.s_min_uncond = self.s_min_uncond if self.s_min_uncond is not None else opts.s_min_uncond -- cgit v1.2.3 From 353c876172a48c5044130249370c9303e611dd8b Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 14 Aug 2023 10:43:18 +0300 Subject: fix API always using -1 as seed --- modules/api/api.py | 2 ++ modules/processing.py | 4 +++- modules/processing_scripts/refiner.py | 2 +- modules/processing_scripts/seed.py | 2 +- modules/scripts.py | 12 +++++++++--- 5 files changed, 16 insertions(+), 6 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/api/api.py b/modules/api/api.py index 908c4514..fb2c2ce9 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -330,6 +330,7 @@ class Api: with self.queue_lock: with closing(StableDiffusionProcessingTxt2Img(sd_model=shared.sd_model, **args)) as p: + p.is_api = True p.scripts = script_runner p.outpath_grids = opts.outdir_txt2img_grids p.outpath_samples = opts.outdir_txt2img_samples @@ -390,6 +391,7 @@ class Api: with self.queue_lock: with closing(StableDiffusionProcessingImg2Img(sd_model=shared.sd_model, **args)) as p: p.init_images = [decode_base64_to_image(x) for x in init_images] + p.is_api = True p.scripts = script_runner p.outpath_grids = opts.outdir_img2img_grids p.outpath_samples = opts.outdir_img2img_samples diff --git a/modules/processing.py b/modules/processing.py index f34ba48a..b1eac2ab 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -194,6 +194,8 @@ class StableDiffusionProcessing: sd_vae_name: str = field(default=None, init=False) sd_vae_hash: str = field(default=None, init=False) + is_api: bool = field(default=False, init=False) + def __post_init__(self): if self.sampler_index is not None: print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr) @@ -258,7 +260,7 @@ class StableDiffusionProcessing: def setup_scripts(self): self.scripts_setup_complete = True - self.scripts.setup_scrips(self) + self.scripts.setup_scrips(self, is_ui=not self.is_api) def comment(self, text): self.comments[text] = 1 diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index 3c5b37d2..b389c4ef 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -5,7 +5,7 @@ from modules.ui_common import create_refresh_button from modules.ui_components import InputAccordion -class ScriptRefiner(scripts.Script): +class ScriptRefiner(scripts.ScriptBuiltinUI): section = "accordions" create_group = False diff --git a/modules/processing_scripts/seed.py b/modules/processing_scripts/seed.py index 6ce3b2fc..6b6ff987 100644 --- a/modules/processing_scripts/seed.py +++ b/modules/processing_scripts/seed.py @@ -7,7 +7,7 @@ from modules.shared import cmd_opts from modules.ui_components import ToolButton -class ScriptSeed(scripts.ScriptBuiltin): +class ScriptSeed(scripts.ScriptBuiltinUI): section = "seed" create_group = False diff --git a/modules/scripts.py b/modules/scripts.py index cbdac2b5..fcab5d3a 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -68,6 +68,9 @@ class Script: on_after_component_elem_id = None """list of callbacks to be called after a component with an elem_id is created""" + setup_for_ui_only = False + """If true, the script setup will only be run in Gradio UI, not in API""" + def title(self): """this function should return the title of the script. This is what will be displayed in the dropdown menu.""" @@ -258,7 +261,6 @@ class Script: self.on_after_component_elem_id.append((elem_id, callback)) - def describe(self): """unused""" return "" @@ -280,7 +282,8 @@ class Script: pass -class ScriptBuiltin(Script): +class ScriptBuiltinUI(Script): + setup_for_ui_only = True def elem_id(self, item_id): """helper function to generate id for a HTML element, constructs final id out of tab and user-supplied item_id""" @@ -728,8 +731,11 @@ class ScriptRunner: except Exception: errors.report(f"Error running before_hr: {script.filename}", exc_info=True) - def setup_scrips(self, p): + def setup_scrips(self, p, *, is_ui=True): for script in self.alwayson_scripts: + if not is_ui and script.setup_for_ui_only: + continue + try: script_args = p.script_args[script.args_from:script.args_to] script.setup(p, *script_args) -- cgit v1.2.3 From 99ab3d43a71e3f66e57d3cd2013b97c97e7ab69b Mon Sep 17 00:00:00 2001 From: Robert Barron Date: Mon, 14 Aug 2023 00:41:36 -0700 Subject: hires prompt timeline: merge to latests, slightly simplify diff --- modules/processing.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 6d3c9365..75f1d66f 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -394,7 +394,7 @@ class StableDiffusionProcessing: self.main_prompt = self.all_prompts[0] self.main_negative_prompt = self.all_negative_prompts[0] - def cached_params(self, required_prompts, steps, hires_steps, extra_network_data, use_old_scheduling): + def cached_params(self, required_prompts, steps, extra_network_data, hires_steps=None, use_old_scheduling=False): """Returns parameters that invalidate the cond cache if changed""" return ( @@ -424,7 +424,7 @@ class StableDiffusionProcessing: caches is a list with items described above. """ - cached_params = self.cached_params(required_prompts, steps, hires_steps, extra_network_data, shared.opts.use_old_scheduling) + cached_params = self.cached_params(required_prompts, steps, extra_network_data, hires_steps, shared.opts.use_old_scheduling) for cache in caches: if cache[0] is not None and cached_params == cache[0]: @@ -447,8 +447,8 @@ class StableDiffusionProcessing: self.step_multiplier = total_steps // self.steps self.firstpass_steps = total_steps - self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, total_steps, [self.cached_uc], self.extra_network_data, None) - self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, total_steps, [self.cached_c], self.extra_network_data, None ) + self.uc = self.get_conds_with_caching(prompt_parser.get_learned_conditioning, negative_prompts, total_steps, [self.cached_uc], self.extra_network_data) + self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, total_steps, [self.cached_c], self.extra_network_data) def get_conds(self): return self.c, self.uc -- cgit v1.2.3 From f23e5ce2daad86d891a5ecdf3f3bad43d7a09c1a Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 14 Aug 2023 17:58:54 +0300 Subject: revert changed inpainting mask conditioning calculation after #12311 --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index b1eac2ab..c983d001 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -1499,7 +1499,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): elif self.inpainting_fill == 3: self.init_latent = self.init_latent * self.mask - self.image_conditioning = self.img2img_image_conditioning(image, self.init_latent, image_mask) + self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, self.init_latent, image_mask) def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): x = self.rng.next() -- cgit v1.2.3 From 7e77a38cbc155baf96db66064afbdd40df715335 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 15 Aug 2023 08:27:50 +0300 Subject: get XYZ plot to work with recent changes to refined specified in fields of p rather than in settings --- modules/processing.py | 2 +- modules/processing_scripts/refiner.py | 2 +- scripts/xyz_grid.py | 18 +++++------------- 3 files changed, 7 insertions(+), 15 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index c983d001..25a19a77 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -746,7 +746,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.tiling is None: p.tiling = opts.tiling - if p.refiner_checkpoint not in (None, "", "None"): + if p.refiner_checkpoint not in (None, "", "None", "none"): p.refiner_checkpoint_info = sd_models.get_closet_checkpoint_match(p.refiner_checkpoint) if p.refiner_checkpoint_info is None: raise Exception(f'Could not find checkpoint with name {p.refiner_checkpoint}') diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index b389c4ef..29ccb78f 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -42,7 +42,7 @@ class ScriptRefiner(scripts.ScriptBuiltinUI): # the actual implementation is in sd_samplers_common.py, apply_refiner if not enable_refiner or refiner_checkpoint in (None, "", "None"): - p.refiner_checkpoint_info = None + p.refiner_checkpoint = None p.refiner_switch_at = None else: p.refiner_checkpoint = refiner_checkpoint diff --git a/scripts/xyz_grid.py b/scripts/xyz_grid.py index 017a470f..2217cc69 100644 --- a/scripts/xyz_grid.py +++ b/scripts/xyz_grid.py @@ -85,20 +85,12 @@ def confirm_checkpoints(p, xs): if modules.sd_models.get_closet_checkpoint_match(x) is None: raise RuntimeError(f"Unknown checkpoint: {x}") -def apply_refiner_checkpoint(p, x, xs): - if x == 'None': - p.override_settings['sd_refiner_checkpoint'] = 'None' - return - info = modules.sd_models.get_closet_checkpoint_match(x) - if info is None: - raise RuntimeError(f"Unknown checkpoint: {x}") - p.override_settings['sd_refiner_checkpoint'] = info.name - -def confirm_refiner_checkpoints(p, xs): +def confirm_checkpoints_or_none(p, xs): for x in xs: - if x == 'None': + if x in (None, "", "None", "none"): continue + if modules.sd_models.get_closet_checkpoint_match(x) is None: raise RuntimeError(f"Unknown checkpoint: {x}") @@ -267,8 +259,8 @@ axis_options = [ AxisOption("Token merging ratio", float, apply_override('token_merging_ratio')), AxisOption("Token merging ratio high-res", float, apply_override('token_merging_ratio_hr')), AxisOption("Always discard next-to-last sigma", str, apply_override('always_discard_next_to_last_sigma', boolean=True), choices=boolean_choice(reverse=True)), - AxisOption("Refiner checkpoint", str, apply_refiner_checkpoint, format_value=format_remove_path, confirm=confirm_refiner_checkpoints, cost=1.0, choices=lambda: ['None'] + sorted(sd_models.checkpoints_list, key=str.casefold)), - AxisOption("Refiner switch at", float, apply_override('sd_refiner_switch_at')) + AxisOption("Refiner checkpoint", str, apply_field('refiner_checkpoint'), format_value=format_remove_path, confirm=confirm_checkpoints_or_none, cost=1.0, choices=lambda: ['None'] + sorted(sd_models.checkpoints_list, key=str.casefold)), + AxisOption("Refiner switch at", float, apply_field('refiner_switch_at')), ] -- cgit v1.2.3 From 79d4e81984171a047d6c71d97a67dda7dd87c43c Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 15 Aug 2023 08:46:17 +0300 Subject: fix processing error that happens if batch_size is not a multiple of how many prompts/negative prompts there are #12509 --- modules/processing.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 25a19a77..1d098302 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -382,13 +382,18 @@ class StableDiffusionProcessing: def setup_prompts(self): if type(self.prompt) == list: self.all_prompts = self.prompt + elif type(self.negative_prompt) == list: + self.all_prompts = [self.prompt] * len(self.negative_prompt) else: self.all_prompts = self.batch_size * self.n_iter * [self.prompt] if type(self.negative_prompt) == list: self.all_negative_prompts = self.negative_prompt else: - self.all_negative_prompts = self.batch_size * self.n_iter * [self.negative_prompt] + self.all_negative_prompts = [self.negative_prompt] * len(self.all_prompts) + + if len(self.all_prompts) != len(self.all_negative_prompts): + raise RuntimeError(f"Received a different number of prompts ({len(self.all_prompts)}) and negative prompts ({len(self.all_negative_prompts)})") self.all_prompts = [shared.prompt_styles.apply_styles_to_prompt(x, self.styles) for x in self.all_prompts] self.all_negative_prompts = [shared.prompt_styles.apply_negative_styles_to_prompt(x, self.styles) for x in self.all_negative_prompts] -- cgit v1.2.3 From 0f77139253f5481d62f7c1eddc220355440b2d1f Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Tue, 15 Aug 2023 14:24:55 -0400 Subject: Fix inpaint upload for alpha masks, create reusable function --- modules/img2img.py | 2 +- modules/processing.py | 10 +++++++++- modules/ui.py | 2 +- 3 files changed, 11 insertions(+), 3 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/img2img.py b/modules/img2img.py index ac9fd3f8..328cb0e9 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -129,7 +129,7 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s mask = None elif mode == 2: # inpaint image, mask = init_img_with_mask["image"], init_img_with_mask["mask"] - mask = mask.split()[-1].convert("L").point(lambda x: 255 if x > 128 else 0) + mask = processing.create_binary_mask(mask) image = image.convert("RGB") elif mode == 3: # inpaint sketch image = inpaint_color_sketch diff --git a/modules/processing.py b/modules/processing.py index 1d098302..e62db62f 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -81,6 +81,12 @@ def apply_overlay(image, paste_loc, index, overlays): return image +def create_binary_mask(image): + if image.mode == 'RGBA' and image.getextrema()[-1] != (255, 255): + image = image.split()[-1].convert("L").point(lambda x: 255 if x > 128 else 0) + else: + image = image.convert('L') + return image def txt2img_image_conditioning(sd_model, x, width, height): if sd_model.model.conditioning_key in {'hybrid', 'concat'}: # Inpainting models @@ -1385,7 +1391,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image_mask = self.image_mask if image_mask is not None: - image_mask = image_mask.convert('L') + # image_mask is passed in as RGBA by Gradio to support alpha masks, + # but we still want to support binary masks. + image_mask = create_binary_mask(image_mask) if self.inpainting_mask_invert: image_mask = ImageOps.invert(image_mask) diff --git a/modules/ui.py b/modules/ui.py index a6b1f964..c98d9849 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -598,7 +598,7 @@ def create_ui(): with gr.TabItem('Inpaint upload', id='inpaint_upload', elem_id="img2img_inpaint_upload_tab") as tab_inpaint_upload: init_img_inpaint = gr.Image(label="Image for img2img", show_label=False, source="upload", interactive=True, type="pil", elem_id="img_inpaint_base") - init_mask_inpaint = gr.Image(label="Mask", source="upload", interactive=True, type="pil", elem_id="img_inpaint_mask") + init_mask_inpaint = gr.Image(label="Mask", source="upload", interactive=True, type="pil", image_mode="RGBA", elem_id="img_inpaint_mask") with gr.TabItem('Batch', id='batch', elem_id="img2img_batch_tab") as tab_batch: hidden = '
Disabled when launched with --hide-ui-dir-config.' if shared.cmd_opts.hide_ui_dir_config else '' -- cgit v1.2.3 From 640cb1bb8dfbd05a3927753090bc80fb5b9737b1 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Fri, 18 Aug 2023 17:14:02 +0900 Subject: fix model override logic do not need extra logic to unload refine model --- modules/processing.py | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index e62db62f..263baefc 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -702,11 +702,8 @@ def process_images(p: StableDiffusionProcessing) -> Processed: stored_opts = {k: opts.data[k] for k in p.override_settings.keys()} try: - # after running refiner, the refiner model is not unloaded - webui swaps back to main model here - if shared.sd_model.sd_checkpoint_info.title != opts.sd_model_checkpoint: - sd_models.reload_model_weights() - # if no checkpoint override or the override checkpoint can't be found, remove override entry and load opts checkpoint + # and if after running refiner, the refiner model is not unloaded - webui swaps back to main model here, if model over is present it will be reloaded afterwards if sd_models.checkpoint_aliases.get(p.override_settings.get('sd_model_checkpoint')) is None: p.override_settings.pop('sd_model_checkpoint', None) sd_models.reload_model_weights() -- cgit v1.2.3 From 9d1d63afcaffc57e9dff5bdc59643ddd60f377e4 Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Fri, 18 Aug 2023 05:55:10 -0400 Subject: Exit out of hires fix if interrupted earlier --- modules/processing.py | 3 +++ 1 file changed, 3 insertions(+) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index e62db62f..b4cb5a82 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -1155,6 +1155,9 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): devices.torch_gc() def sample_hr_pass(self, samples, decoded_samples, seeds, subseeds, subseed_strength, prompts): + if shared.state.interrupted: + return samples + self.is_hr_pass = True target_width = self.hr_upscale_to_x -- cgit v1.2.3 From 61c1261e4e50385aab68b84c0f886911466044bb Mon Sep 17 00:00:00 2001 From: XDOneDude <106700244+XDOneDude@users.noreply.github.com> Date: Fri, 18 Aug 2023 21:56:15 -0400 Subject: more grammar fixes --- modules/processing.py | 24 ++++++++++++------------ modules/prompt_parser.py | 2 +- 2 files changed, 13 insertions(+), 13 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index e62db62f..0315e1fd 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -386,14 +386,14 @@ class StableDiffusionProcessing: return self.token_merging_ratio or opts.token_merging_ratio def setup_prompts(self): - if type(self.prompt) == list: + if isinstance(self.prompt,list): self.all_prompts = self.prompt - elif type(self.negative_prompt) == list: + elif isinstance(self.negative_prompt, list): self.all_prompts = [self.prompt] * len(self.negative_prompt) else: self.all_prompts = self.batch_size * self.n_iter * [self.prompt] - if type(self.negative_prompt) == list: + if isinstance(self.negative_prompt, list): self.all_negative_prompts = self.negative_prompt else: self.all_negative_prompts = [self.negative_prompt] * len(self.all_prompts) @@ -512,10 +512,10 @@ class Processed: self.s_noise = p.s_noise self.s_min_uncond = p.s_min_uncond self.sampler_noise_scheduler_override = p.sampler_noise_scheduler_override - self.prompt = self.prompt if type(self.prompt) != list else self.prompt[0] - self.negative_prompt = self.negative_prompt if type(self.negative_prompt) != list else self.negative_prompt[0] - self.seed = int(self.seed if type(self.seed) != list else self.seed[0]) if self.seed is not None else -1 - self.subseed = int(self.subseed if type(self.subseed) != list else self.subseed[0]) if self.subseed is not None else -1 + self.prompt = self.prompt if not isinstance(self.prompt, list) else self.prompt[0] + self.negative_prompt = self.negative_prompt if not isinstance(self.negative_prompt, list) else self.negative_prompt[0] + self.seed = int(self.seed if not isinstance(self.seed, list) else self.seed[0]) if self.seed is not None else -1 + self.subseed = int(self.subseed if not isinstance(self.subseed, list) else self.subseed[0]) if self.subseed is not None else -1 self.is_using_inpainting_conditioning = p.is_using_inpainting_conditioning self.all_prompts = all_prompts or p.all_prompts or [self.prompt] @@ -741,7 +741,7 @@ def process_images(p: StableDiffusionProcessing) -> Processed: def process_images_inner(p: StableDiffusionProcessing) -> Processed: """this is the main loop that both txt2img and img2img use; it calls func_init once inside all the scopes and func_sample once per batch""" - if type(p.prompt) == list: + if isinstance(p.prompt, list): assert(len(p.prompt) > 0) else: assert p.prompt is not None @@ -772,12 +772,12 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.setup_prompts() - if type(seed) == list: + if isinstance(seed, list): p.all_seeds = seed else: p.all_seeds = [int(seed) + (x if p.subseed_strength == 0 else 0) for x in range(len(p.all_prompts))] - if type(subseed) == list: + if isinstance(subseed, list): p.all_subseeds = subseed else: p.all_subseeds = [int(subseed) + x for x in range(len(p.all_prompts))] @@ -1268,12 +1268,12 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): if self.hr_negative_prompt == '': self.hr_negative_prompt = self.negative_prompt - if type(self.hr_prompt) == list: + if isinstance(self.hr_prompt, list): self.all_hr_prompts = self.hr_prompt else: self.all_hr_prompts = self.batch_size * self.n_iter * [self.hr_prompt] - if type(self.hr_negative_prompt) == list: + if isinstance(self.hr_negative_prompt, list): self.all_hr_negative_prompts = self.hr_negative_prompt else: self.all_hr_negative_prompts = self.batch_size * self.n_iter * [self.hr_negative_prompt] diff --git a/modules/prompt_parser.py b/modules/prompt_parser.py index 32d214e3..e811ae99 100644 --- a/modules/prompt_parser.py +++ b/modules/prompt_parser.py @@ -86,7 +86,7 @@ def get_learned_conditioning_prompt_schedules(prompts, steps): yield args[(step - 1) % len(args)] def start(self, args): def flatten(x): - if type(x) == str: + if isinstance(x, str): yield x else: for gen in x: -- cgit v1.2.3 From d02c4da483a656331f641224fe5da5f694580066 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 21 Aug 2023 08:58:15 +0300 Subject: also prevent changing API options via override_settings --- modules/options.py | 4 ++-- modules/processing.py | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'modules/processing.py') diff --git a/modules/options.py b/modules/options.py index 41d1b672..758b1ce5 100644 --- a/modules/options.py +++ b/modules/options.py @@ -116,7 +116,7 @@ class Options: return super(Options, self).__getattribute__(item) - def set(self, key, value, is_api=False): + def set(self, key, value, is_api=False, run_callbacks=True): """sets an option and calls its onchange callback, returning True if the option changed and False otherwise""" oldval = self.data.get(key, None) @@ -135,7 +135,7 @@ class Options: except RuntimeError: return False - if option.onchange is not None: + if run_callbacks and option.onchange is not None: try: option.onchange() except Exception as e: diff --git a/modules/processing.py b/modules/processing.py index d4926524..e60cc92b 100755 --- a/modules/processing.py +++ b/modules/processing.py @@ -709,7 +709,7 @@ def process_images(p: StableDiffusionProcessing) -> Processed: sd_models.reload_model_weights() for k, v in p.override_settings.items(): - setattr(opts, k, v) + opts.set(k, v, is_api=True, run_callbacks=False) if k == 'sd_model_checkpoint': sd_models.reload_model_weights() -- cgit v1.2.3 From 995ff5902fe0567e4cb2aa2e8ac3d554fca7b1ab Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 24 Aug 2023 10:07:54 +0300 Subject: add infotext for use_old_scheduling option --- modules/processing.py | 6 ++++++ modules/shared_options.py | 2 +- 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'modules/processing.py') diff --git a/modules/processing.py b/modules/processing.py index 066351c1..7dc931ba 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -437,6 +437,12 @@ class StableDiffusionProcessing: caches is a list with items described above. """ + if shared.opts.use_old_scheduling: + old_schedules = prompt_parser.get_learned_conditioning_prompt_schedules(required_prompts, steps, hires_steps, False) + new_schedules = prompt_parser.get_learned_conditioning_prompt_schedules(required_prompts, steps, hires_steps, True) + if old_schedules != new_schedules: + self.extra_generation_params["Old prompt editing timelines"] = True + cached_params = self.cached_params(required_prompts, steps, extra_network_data, hires_steps, shared.opts.use_old_scheduling) for cache in caches: diff --git a/modules/shared_options.py b/modules/shared_options.py index d1389838..83f56314 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -203,7 +203,7 @@ options_templates.update(options_section(('compatibility', "Compatibility"), { "use_old_hires_fix_width_height": OptionInfo(False, "For hires fix, use width/height sliders to set final resolution rather than first pass (disables Upscale by, Resize width/height to)."), "dont_fix_second_order_samplers_schedule": OptionInfo(False, "Do not fix prompt schedule for second order samplers."), "hires_fix_use_firstpass_conds": OptionInfo(False, "For hires fix, calculate conds of second pass using extra networks of first pass."), - "use_old_scheduling": OptionInfo(False, "Use old prompt where first pass and hires both used the same timeline, and < 1 meant relative and >= 1 meant absolute"), + "use_old_scheduling": OptionInfo(False, "Use old prompt editing timelines.", infotext="Old prompt editing timelines").info("For [red:green:N]; old: If N < 1, it's a fraction of steps (and hires fix uses range from 0 to 1), if N >= 1, it's an absolute number of steps; new: If N has a decimal point in it, it's a fraction of steps (and hires fix uses range from 1 to 2), othewrwise it's an absolute number of steps"), })) options_templates.update(options_section(('interrogate', "Interrogate"), { -- cgit v1.2.3