From 75c4511e6b81ae8fb0dbd932043e8eb35cd09f72 Mon Sep 17 00:00:00 2001 From: zhaohu xing <920232796@qq.com> Date: Tue, 29 Nov 2022 10:28:41 +0800 Subject: add AltDiffusion to webui Signed-off-by: zhaohu xing <920232796@qq.com> --- modules/shared.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index c93ae2a3..9941d2f4 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -21,7 +21,7 @@ from modules.paths import models_path, script_path, sd_path sd_model_file = os.path.join(script_path, 'model.ckpt') default_sd_model_file = sd_model_file parser = argparse.ArgumentParser() -parser.add_argument("--config", type=str, default=os.path.join(sd_path, "configs/stable-diffusion/v1-inference.yaml"), help="path to config which constructs model",) +parser.add_argument("--config", type=str, default="configs/altdiffusion/ad-inference.yaml", help="path to config which constructs model",) parser.add_argument("--ckpt", type=str, default=sd_model_file, help="path to checkpoint of stable diffusion model; if specified, this checkpoint will be added to the list of checkpoints and loaded",) parser.add_argument("--ckpt-dir", type=str, default=None, help="Path to directory with stable diffusion checkpoints") parser.add_argument("--gfpgan-dir", type=str, help="GFPGAN directory", default=('./src/gfpgan' if os.path.exists('./src/gfpgan') else './GFPGAN')) @@ -106,6 +106,10 @@ restricted_opts = { "outdir_txt2img_grids", "outdir_save", } +from omegaconf import OmegaConf +config = OmegaConf.load(f"{cmd_opts.config}") +# XLMR-Large +text_model_name = config.model.params.cond_stage_config.params.name cmd_opts.disable_extension_access = (cmd_opts.share or cmd_opts.listen or cmd_opts.server_name) and not cmd_opts.enable_insecure_extension_access -- cgit v1.2.3 From 9c86fb8cace6d8ac0843e0ddad0ba5ae7f3148c9 Mon Sep 17 00:00:00 2001 From: zhaohu xing <920232796@qq.com> Date: Fri, 2 Dec 2022 16:08:46 +0800 Subject: fix bug Signed-off-by: zhaohu xing <920232796@qq.com> --- modules/shared.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index 1408dee3..ac7678c3 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -111,7 +111,11 @@ restricted_opts = { from omegaconf import OmegaConf config = OmegaConf.load(f"{cmd_opts.config}") # XLMR-Large -text_model_name = config.model.params.cond_stage_config.params.name +try: + text_model_name = config.model.params.cond_stage_config.params.name + +except : + text_model_name = "stable_diffusion" cmd_opts.disable_extension_access = (cmd_opts.share or cmd_opts.listen or cmd_opts.server_name) and not cmd_opts.enable_insecure_extension_access -- cgit v1.2.3 From 965fc5ac5a6ccdf38342e21c97183011a04e799e Mon Sep 17 00:00:00 2001 From: zhaohu xing <920232796@qq.com> Date: Tue, 6 Dec 2022 16:15:15 +0800 Subject: delete a file Signed-off-by: zhaohu xing <920232796@qq.com> --- .DS_Store | Bin 6148 -> 0 bytes modules/shared.py | 2 +- 2 files changed, 1 insertion(+), 1 deletion(-) delete mode 100644 .DS_Store (limited to 'modules/shared.py') diff --git a/.DS_Store b/.DS_Store deleted file mode 100644 index 5008ddfc..00000000 Binary files a/.DS_Store and /dev/null differ diff --git a/modules/shared.py b/modules/shared.py index 522c56c1..8419b531 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -22,7 +22,7 @@ demo = None sd_model_file = os.path.join(script_path, 'model.ckpt') default_sd_model_file = sd_model_file parser = argparse.ArgumentParser() -parser.add_argument("--config", type=str, default="configs/altdiffusion/ad-inference.yaml", help="path to config which constructs model",) +parser.add_argument("--config", type=str, default=os.path.join(script_path, "v1-inference.yaml"), help="path to config which constructs model",) parser.add_argument("--ckpt", type=str, default=sd_model_file, help="path to checkpoint of stable diffusion model; if specified, this checkpoint will be added to the list of checkpoints and loaded",) parser.add_argument("--ckpt-dir", type=str, default=None, help="Path to directory with stable diffusion checkpoints") parser.add_argument("--gfpgan-dir", type=str, help="GFPGAN directory", default=('./src/gfpgan' if os.path.exists('./src/gfpgan') else './GFPGAN')) -- cgit v1.2.3 From f23a822f1c9cb3bd2e8772c75af429e06515eaef Mon Sep 17 00:00:00 2001 From: Philpax Date: Sat, 24 Dec 2022 20:45:16 +1100 Subject: feat(api): include job_timestamp in progress --- modules/shared.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index 8ea3b441..f356dbf7 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -171,6 +171,7 @@ class State: "interrupted": self.skipped, "job": self.job, "job_count": self.job_count, + "job_timestamp": self.job_timestamp, "job_no": self.job_no, "sampling_step": self.sampling_step, "sampling_steps": self.sampling_steps, -- cgit v1.2.3 From 893933e05ad267778111b4fad6d1ecb80937afdf Mon Sep 17 00:00:00 2001 From: hitomi Date: Sun, 25 Dec 2022 20:49:25 +0800 Subject: Add memory cache for VAE weights --- modules/sd_vae.py | 31 +++++++++++++++++++++++++------ modules/shared.py | 1 + 2 files changed, 26 insertions(+), 6 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/sd_vae.py b/modules/sd_vae.py index 3856418e..ac71d62d 100644 --- a/modules/sd_vae.py +++ b/modules/sd_vae.py @@ -1,5 +1,6 @@ import torch import os +import collections from collections import namedtuple from modules import shared, devices, script_callbacks from modules.paths import models_path @@ -30,6 +31,7 @@ base_vae = None loaded_vae_file = None checkpoint_info = None +checkpoints_loaded = collections.OrderedDict() def get_base_vae(model): if base_vae is not None and checkpoint_info == model.sd_checkpoint_info and model: @@ -149,13 +151,30 @@ def load_vae(model, vae_file=None): global first_load, vae_dict, vae_list, loaded_vae_file # save_settings = False + cache_enabled = shared.opts.sd_vae_checkpoint_cache > 0 + if vae_file: - assert os.path.isfile(vae_file), f"VAE file doesn't exist: {vae_file}" - print(f"Loading VAE weights from: {vae_file}") - store_base_vae(model) - vae_ckpt = torch.load(vae_file, map_location=shared.weight_load_location) - vae_dict_1 = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss" and k not in vae_ignore_keys} - _load_vae_dict(model, vae_dict_1) + if cache_enabled and vae_file in checkpoints_loaded: + # use vae checkpoint cache + print(f"Loading VAE weights [{get_filename(vae_file)}] from cache") + store_base_vae(model) + _load_vae_dict(model, checkpoints_loaded[vae_file]) + else: + assert os.path.isfile(vae_file), f"VAE file doesn't exist: {vae_file}" + print(f"Loading VAE weights from: {vae_file}") + store_base_vae(model) + vae_ckpt = torch.load(vae_file, map_location=shared.weight_load_location) + vae_dict_1 = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss" and k not in vae_ignore_keys} + _load_vae_dict(model, vae_dict_1) + + if cache_enabled: + # cache newly loaded vae + checkpoints_loaded[vae_file] = vae_dict_1.copy() + + # clean up cache if limit is reached + if cache_enabled: + while len(checkpoints_loaded) > shared.opts.sd_vae_checkpoint_cache + 1: # we need to count the current model + checkpoints_loaded.popitem(last=False) # LRU # If vae used is not in dict, update it # It will be removed on refresh though diff --git a/modules/shared.py b/modules/shared.py index d4ddeea0..671d30e1 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -356,6 +356,7 @@ options_templates.update(options_section(('training', "Training"), { options_templates.update(options_section(('sd', "Stable Diffusion"), { "sd_model_checkpoint": OptionInfo(None, "Stable Diffusion checkpoint", gr.Dropdown, lambda: {"choices": list_checkpoint_tiles()}, refresh=refresh_checkpoints), "sd_checkpoint_cache": OptionInfo(0, "Checkpoints to cache in RAM", gr.Slider, {"minimum": 0, "maximum": 10, "step": 1}), + "sd_vae_checkpoint_cache": OptionInfo(0, "VAE Checkpoints to cache in RAM", gr.Slider, {"minimum": 0, "maximum": 10, "step": 1}), "sd_vae": OptionInfo("auto", "SD VAE", gr.Dropdown, lambda: {"choices": sd_vae.vae_list}, refresh=sd_vae.refresh_vae_list), "sd_vae_as_default": OptionInfo(False, "Ignore selected VAE for stable diffusion checkpoints that have their own .vae.pt next to them"), "sd_hypernetwork": OptionInfo("None", "Hypernetwork", gr.Dropdown, lambda: {"choices": ["None"] + [x for x in hypernetworks.keys()]}, refresh=reload_hypernetworks), -- cgit v1.2.3 From 463048344fc036b262aa132584b65ee6e9fec6cf Mon Sep 17 00:00:00 2001 From: Vladimir Mandic Date: Fri, 30 Dec 2022 19:41:47 -0500 Subject: fix shared state dictionary --- modules/shared.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index d4ddeea0..9a13fb60 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -168,7 +168,7 @@ class State: def dict(self): obj = { "skipped": self.skipped, - "interrupted": self.skipped, + "interrupted": self.interrupted, "job": self.job, "job_count": self.job_count, "job_no": self.job_no, -- cgit v1.2.3 From f34c7341720fb2059992926c9f9ae6ff25f7385b Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 31 Dec 2022 18:06:35 +0300 Subject: alt-diffusion integration --- configs/alt-diffusion-inference.yaml | 72 ++++++++++++++++++++++++++++++++++ configs/altdiffusion/ad-inference.yaml | 72 ---------------------------------- configs/v1-inference.yaml | 70 +++++++++++++++++++++++++++++++++ modules/sd_hijack.py | 18 +++++---- modules/sd_hijack_clip.py | 14 +++---- modules/sd_hijack_xlmr.py | 34 ++++++++++++++++ modules/shared.py | 10 +---- v1-inference.yaml | 70 --------------------------------- 8 files changed, 192 insertions(+), 168 deletions(-) create mode 100644 configs/alt-diffusion-inference.yaml delete mode 100644 configs/altdiffusion/ad-inference.yaml create mode 100644 configs/v1-inference.yaml create mode 100644 modules/sd_hijack_xlmr.py delete mode 100644 v1-inference.yaml (limited to 'modules/shared.py') diff --git a/configs/alt-diffusion-inference.yaml b/configs/alt-diffusion-inference.yaml new file mode 100644 index 00000000..cfbee72d --- /dev/null +++ b/configs/alt-diffusion-inference.yaml @@ -0,0 +1,72 @@ +model: + base_learning_rate: 1.0e-04 + target: ldm.models.diffusion.ddpm.LatentDiffusion + params: + linear_start: 0.00085 + linear_end: 0.0120 + num_timesteps_cond: 1 + log_every_t: 200 + timesteps: 1000 + first_stage_key: "jpg" + cond_stage_key: "txt" + image_size: 64 + channels: 4 + cond_stage_trainable: false # Note: different from the one we trained before + conditioning_key: crossattn + monitor: val/loss_simple_ema + scale_factor: 0.18215 + use_ema: False + + scheduler_config: # 10000 warmup steps + target: ldm.lr_scheduler.LambdaLinearScheduler + params: + warm_up_steps: [ 10000 ] + cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases + f_start: [ 1.e-6 ] + f_max: [ 1. ] + f_min: [ 1. ] + + unet_config: + target: ldm.modules.diffusionmodules.openaimodel.UNetModel + params: + image_size: 32 # unused + in_channels: 4 + out_channels: 4 + model_channels: 320 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + first_stage_config: + target: ldm.models.autoencoder.AutoencoderKL + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity + + cond_stage_config: + target: modules.xlmr.BertSeriesModelWithTransformation + params: + name: "XLMR-Large" \ No newline at end of file diff --git a/configs/altdiffusion/ad-inference.yaml b/configs/altdiffusion/ad-inference.yaml deleted file mode 100644 index cfbee72d..00000000 --- a/configs/altdiffusion/ad-inference.yaml +++ /dev/null @@ -1,72 +0,0 @@ -model: - base_learning_rate: 1.0e-04 - target: ldm.models.diffusion.ddpm.LatentDiffusion - params: - linear_start: 0.00085 - linear_end: 0.0120 - num_timesteps_cond: 1 - log_every_t: 200 - timesteps: 1000 - first_stage_key: "jpg" - cond_stage_key: "txt" - image_size: 64 - channels: 4 - cond_stage_trainable: false # Note: different from the one we trained before - conditioning_key: crossattn - monitor: val/loss_simple_ema - scale_factor: 0.18215 - use_ema: False - - scheduler_config: # 10000 warmup steps - target: ldm.lr_scheduler.LambdaLinearScheduler - params: - warm_up_steps: [ 10000 ] - cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases - f_start: [ 1.e-6 ] - f_max: [ 1. ] - f_min: [ 1. ] - - unet_config: - target: ldm.modules.diffusionmodules.openaimodel.UNetModel - params: - image_size: 32 # unused - in_channels: 4 - out_channels: 4 - model_channels: 320 - attention_resolutions: [ 4, 2, 1 ] - num_res_blocks: 2 - channel_mult: [ 1, 2, 4, 4 ] - num_heads: 8 - use_spatial_transformer: True - transformer_depth: 1 - context_dim: 768 - use_checkpoint: True - legacy: False - - first_stage_config: - target: ldm.models.autoencoder.AutoencoderKL - params: - embed_dim: 4 - monitor: val/rec_loss - ddconfig: - double_z: true - z_channels: 4 - resolution: 256 - in_channels: 3 - out_ch: 3 - ch: 128 - ch_mult: - - 1 - - 2 - - 4 - - 4 - num_res_blocks: 2 - attn_resolutions: [] - dropout: 0.0 - lossconfig: - target: torch.nn.Identity - - cond_stage_config: - target: modules.xlmr.BertSeriesModelWithTransformation - params: - name: "XLMR-Large" \ No newline at end of file diff --git a/configs/v1-inference.yaml b/configs/v1-inference.yaml new file mode 100644 index 00000000..d4effe56 --- /dev/null +++ b/configs/v1-inference.yaml @@ -0,0 +1,70 @@ +model: + base_learning_rate: 1.0e-04 + target: ldm.models.diffusion.ddpm.LatentDiffusion + params: + linear_start: 0.00085 + linear_end: 0.0120 + num_timesteps_cond: 1 + log_every_t: 200 + timesteps: 1000 + first_stage_key: "jpg" + cond_stage_key: "txt" + image_size: 64 + channels: 4 + cond_stage_trainable: false # Note: different from the one we trained before + conditioning_key: crossattn + monitor: val/loss_simple_ema + scale_factor: 0.18215 + use_ema: False + + scheduler_config: # 10000 warmup steps + target: ldm.lr_scheduler.LambdaLinearScheduler + params: + warm_up_steps: [ 10000 ] + cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases + f_start: [ 1.e-6 ] + f_max: [ 1. ] + f_min: [ 1. ] + + unet_config: + target: ldm.modules.diffusionmodules.openaimodel.UNetModel + params: + image_size: 32 # unused + in_channels: 4 + out_channels: 4 + model_channels: 320 + attention_resolutions: [ 4, 2, 1 ] + num_res_blocks: 2 + channel_mult: [ 1, 2, 4, 4 ] + num_heads: 8 + use_spatial_transformer: True + transformer_depth: 1 + context_dim: 768 + use_checkpoint: True + legacy: False + + first_stage_config: + target: ldm.models.autoencoder.AutoencoderKL + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity + + cond_stage_config: + target: ldm.modules.encoders.modules.FrozenCLIPEmbedder diff --git a/modules/sd_hijack.py b/modules/sd_hijack.py index bce23b03..edcbaf52 100644 --- a/modules/sd_hijack.py +++ b/modules/sd_hijack.py @@ -5,7 +5,7 @@ import modules.textual_inversion.textual_inversion from modules import devices, sd_hijack_optimizations, shared, sd_hijack_checkpoint from modules.hypernetworks import hypernetwork from modules.shared import cmd_opts -from modules import sd_hijack_clip, sd_hijack_open_clip, sd_hijack_unet +from modules import sd_hijack_clip, sd_hijack_open_clip, sd_hijack_unet, sd_hijack_xlmr, xlmr from modules.sd_hijack_optimizations import invokeAI_mps_available @@ -68,6 +68,7 @@ def fix_checkpoint(): ldm.modules.diffusionmodules.openaimodel.ResBlock.forward = sd_hijack_checkpoint.ResBlock_forward ldm.modules.diffusionmodules.openaimodel.AttentionBlock.forward = sd_hijack_checkpoint.AttentionBlock_forward + class StableDiffusionModelHijack: fixes = None comments = [] @@ -79,21 +80,22 @@ class StableDiffusionModelHijack: def hijack(self, m): - if shared.text_model_name == "XLMR-Large": + if type(m.cond_stage_model) == xlmr.BertSeriesModelWithTransformation: model_embeddings = m.cond_stage_model.roberta.embeddings model_embeddings.token_embedding = EmbeddingsWithFixes(model_embeddings.word_embeddings, self) - m.cond_stage_model = sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords(m.cond_stage_model, self) - + m.cond_stage_model = sd_hijack_xlmr.FrozenXLMREmbedderWithCustomWords(m.cond_stage_model, self) + elif type(m.cond_stage_model) == ldm.modules.encoders.modules.FrozenCLIPEmbedder: model_embeddings = m.cond_stage_model.transformer.text_model.embeddings model_embeddings.token_embedding = EmbeddingsWithFixes(model_embeddings.token_embedding, self) m.cond_stage_model = sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords(m.cond_stage_model, self) - apply_optimizations() + elif type(m.cond_stage_model) == ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder: m.cond_stage_model.model.token_embedding = EmbeddingsWithFixes(m.cond_stage_model.model.token_embedding, self) m.cond_stage_model = sd_hijack_open_clip.FrozenOpenCLIPEmbedderWithCustomWords(m.cond_stage_model, self) - apply_optimizations() - + + apply_optimizations() + self.clip = m.cond_stage_model fix_checkpoint() @@ -109,7 +111,7 @@ class StableDiffusionModelHijack: def undo_hijack(self, m): - if shared.text_model_name == "XLMR-Large": + if type(m.cond_stage_model) == xlmr.BertSeriesModelWithTransformation: m.cond_stage_model = m.cond_stage_model.wrapped elif type(m.cond_stage_model) == sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords: diff --git a/modules/sd_hijack_clip.py b/modules/sd_hijack_clip.py index 9ea6e1ce..6ec50cca 100644 --- a/modules/sd_hijack_clip.py +++ b/modules/sd_hijack_clip.py @@ -4,7 +4,6 @@ import torch from modules import prompt_parser, devices from modules.shared import opts -import modules.shared as shared def get_target_prompt_token_count(token_count): return math.ceil(max(token_count, 1) / 75) * 75 @@ -177,9 +176,6 @@ class FrozenCLIPEmbedderWithCustomWordsBase(torch.nn.Module): return batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count def forward(self, text): - if shared.text_model_name == "XLMR-Large": - return self.wrapped.encode(text) - use_old = opts.use_old_emphasis_implementation if use_old: batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count = self.process_text_old(text) @@ -257,13 +253,13 @@ class FrozenCLIPEmbedderWithCustomWords(FrozenCLIPEmbedderWithCustomWordsBase): def __init__(self, wrapped, hijack): super().__init__(wrapped, hijack) self.tokenizer = wrapped.tokenizer - if shared.text_model_name == "XLMR-Large": - self.comma_token = None - else : - self.comma_token = [v for k, v in self.tokenizer.get_vocab().items() if k == ','][0] + + vocab = self.tokenizer.get_vocab() + + self.comma_token = vocab.get(',', None) self.token_mults = {} - tokens_with_parens = [(k, v) for k, v in self.tokenizer.get_vocab().items() if '(' in k or ')' in k or '[' in k or ']' in k] + tokens_with_parens = [(k, v) for k, v in vocab.items() if '(' in k or ')' in k or '[' in k or ']' in k] for text, ident in tokens_with_parens: mult = 1.0 for c in text: diff --git a/modules/sd_hijack_xlmr.py b/modules/sd_hijack_xlmr.py new file mode 100644 index 00000000..4ac51c38 --- /dev/null +++ b/modules/sd_hijack_xlmr.py @@ -0,0 +1,34 @@ +import open_clip.tokenizer +import torch + +from modules import sd_hijack_clip, devices +from modules.shared import opts + + +class FrozenXLMREmbedderWithCustomWords(sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords): + def __init__(self, wrapped, hijack): + super().__init__(wrapped, hijack) + + self.id_start = wrapped.config.bos_token_id + self.id_end = wrapped.config.eos_token_id + self.id_pad = wrapped.config.pad_token_id + + self.comma_token = self.tokenizer.get_vocab().get(',', None) # alt diffusion doesn't have bits for comma + + def encode_with_transformers(self, tokens): + # there's no CLIP Skip here because all hidden layers have size of 1024 and the last one uses a + # trained layer to transform those 1024 into 768 for unet; so you can't choose which transformer + # layer to work with - you have to use the last + + attention_mask = (tokens != self.id_pad).to(device=tokens.device, dtype=torch.int64) + features = self.wrapped(input_ids=tokens, attention_mask=attention_mask) + z = features['projection_state'] + + return z + + def encode_embedding_init_text(self, init_text, nvpt): + embedding_layer = self.wrapped.roberta.embeddings + ids = self.wrapped.tokenizer(init_text, max_length=nvpt, return_tensors="pt", add_special_tokens=False)["input_ids"] + embedded = embedding_layer.token_embedding.wrapped(ids.to(devices.device)).squeeze(0) + + return embedded diff --git a/modules/shared.py b/modules/shared.py index 2b31e717..715b9169 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -23,7 +23,7 @@ demo = None sd_model_file = os.path.join(script_path, 'model.ckpt') default_sd_model_file = sd_model_file parser = argparse.ArgumentParser() -parser.add_argument("--config", type=str, default=os.path.join(script_path, "v1-inference.yaml"), help="path to config which constructs model",) +parser.add_argument("--config", type=str, default=os.path.join(script_path, "configs/v1-inference.yaml"), help="path to config which constructs model",) parser.add_argument("--ckpt", type=str, default=sd_model_file, help="path to checkpoint of stable diffusion model; if specified, this checkpoint will be added to the list of checkpoints and loaded",) parser.add_argument("--ckpt-dir", type=str, default=None, help="Path to directory with stable diffusion checkpoints") parser.add_argument("--gfpgan-dir", type=str, help="GFPGAN directory", default=('./src/gfpgan' if os.path.exists('./src/gfpgan') else './GFPGAN')) @@ -108,14 +108,6 @@ restricted_opts = { "outdir_txt2img_grids", "outdir_save", } -from omegaconf import OmegaConf -config = OmegaConf.load(f"{cmd_opts.config}") -# XLMR-Large -try: - text_model_name = config.model.params.cond_stage_config.params.name - -except : - text_model_name = "stable_diffusion" cmd_opts.disable_extension_access = (cmd_opts.share or cmd_opts.listen or cmd_opts.server_name) and not cmd_opts.enable_insecure_extension_access diff --git a/v1-inference.yaml b/v1-inference.yaml deleted file mode 100644 index d4effe56..00000000 --- a/v1-inference.yaml +++ /dev/null @@ -1,70 +0,0 @@ -model: - base_learning_rate: 1.0e-04 - target: ldm.models.diffusion.ddpm.LatentDiffusion - params: - linear_start: 0.00085 - linear_end: 0.0120 - num_timesteps_cond: 1 - log_every_t: 200 - timesteps: 1000 - first_stage_key: "jpg" - cond_stage_key: "txt" - image_size: 64 - channels: 4 - cond_stage_trainable: false # Note: different from the one we trained before - conditioning_key: crossattn - monitor: val/loss_simple_ema - scale_factor: 0.18215 - use_ema: False - - scheduler_config: # 10000 warmup steps - target: ldm.lr_scheduler.LambdaLinearScheduler - params: - warm_up_steps: [ 10000 ] - cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases - f_start: [ 1.e-6 ] - f_max: [ 1. ] - f_min: [ 1. ] - - unet_config: - target: ldm.modules.diffusionmodules.openaimodel.UNetModel - params: - image_size: 32 # unused - in_channels: 4 - out_channels: 4 - model_channels: 320 - attention_resolutions: [ 4, 2, 1 ] - num_res_blocks: 2 - channel_mult: [ 1, 2, 4, 4 ] - num_heads: 8 - use_spatial_transformer: True - transformer_depth: 1 - context_dim: 768 - use_checkpoint: True - legacy: False - - first_stage_config: - target: ldm.models.autoencoder.AutoencoderKL - params: - embed_dim: 4 - monitor: val/rec_loss - ddconfig: - double_z: true - z_channels: 4 - resolution: 256 - in_channels: 3 - out_ch: 3 - ch: 128 - ch_mult: - - 1 - - 2 - - 4 - - 4 - num_res_blocks: 2 - attn_resolutions: [] - dropout: 0.0 - lossconfig: - target: torch.nn.Identity - - cond_stage_config: - target: ldm.modules.encoders.modules.FrozenCLIPEmbedder -- cgit v1.2.3 From 29a3a7eb13478297bc7093971b48827ab8246f45 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sun, 1 Jan 2023 01:19:10 +0300 Subject: show sampler selection in dropdown, add option selection to revert to old radio group --- modules/shared.py | 1 + modules/ui.py | 22 +++++++++++++++------- 2 files changed, 16 insertions(+), 7 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index 715b9169..948b9542 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -406,6 +406,7 @@ options_templates.update(options_section(('ui', "User interface"), { "js_modal_lightbox": OptionInfo(True, "Enable full page image viewer"), "js_modal_lightbox_initially_zoomed": OptionInfo(True, "Show images zoomed in by default in full page image viewer"), "show_progress_in_title": OptionInfo(True, "Show generation progress in window title."), + "samplers_in_dropdown": OptionInfo(True, "Use dropdown for sampler selection instead of radio group"), 'quicksettings': OptionInfo("sd_model_checkpoint", "Quicksettings list"), 'localization': OptionInfo("None", "Localization (requires restart)", gr.Dropdown, lambda: {"choices": ["None"] + list(localization.localizations.keys())}, refresh=lambda: localization.list_localizations(cmd_opts.localizations_dir)), })) diff --git a/modules/ui.py b/modules/ui.py index 279b5110..c7b8ea5d 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -643,6 +643,19 @@ Requested path was: {f} return result_gallery, generation_info if tabname != "extras" else html_info_x, html_info, html_log +def create_sampler_and_steps_selection(choices, tabname): + if opts.samplers_in_dropdown: + with gr.Row(elem_id=f"sampler_selection_{tabname}"): + sampler_index = gr.Dropdown(label='Sampling method', elem_id=f"{tabname}_sampling", choices=[x.name for x in choices], value=choices[0].name, type="index") + steps = gr.Slider(minimum=1, maximum=150, step=1, elem_id=f"{tabname}_steps", label="Sampling Steps", value=20) + else: + with gr.Group(elem_id=f"sampler_selection_{tabname}"): + steps = gr.Slider(minimum=1, maximum=150, step=1, elem_id=f"{tabname}_steps", label="Sampling Steps", value=20) + sampler_index = gr.Radio(label='Sampling method', elem_id=f"{tabname}_sampling", choices=[x.name for x in choices], value=choices[0].name, type="index") + + return steps, sampler_index + + def create_ui(): import modules.img2img import modules.txt2img @@ -660,9 +673,6 @@ def create_ui(): dummy_component = gr.Label(visible=False) txt_prompt_img = gr.File(label="", elem_id="txt2img_prompt_image", file_count="single", type="bytes", visible=False) - - - with gr.Row(elem_id='txt2img_progress_row'): with gr.Column(scale=1): pass @@ -674,8 +684,7 @@ def create_ui(): with gr.Row().style(equal_height=False): with gr.Column(variant='panel', elem_id="txt2img_settings"): - steps = gr.Slider(minimum=1, maximum=150, step=1, label="Sampling Steps", value=20) - sampler_index = gr.Radio(label='Sampling method', elem_id="txt2img_sampling", choices=[x.name for x in samplers], value=samplers[0].name, type="index") + steps, sampler_index = create_sampler_and_steps_selection(samplers, "txt2img") with gr.Group(): width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512) @@ -875,8 +884,7 @@ def create_ui(): with gr.Row(): resize_mode = gr.Radio(label="Resize mode", elem_id="resize_mode", show_label=False, choices=["Just resize", "Crop and resize", "Resize and fill", "Just resize (latent upscale)"], type="index", value="Just resize") - steps = gr.Slider(minimum=1, maximum=150, step=1, label="Sampling Steps", value=20) - sampler_index = gr.Radio(label='Sampling method', choices=[x.name for x in samplers_for_img2img], value=samplers_for_img2img[0].name, type="index") + steps, sampler_index = create_sampler_and_steps_selection(samplers_for_img2img, "img2img") with gr.Group(): width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="img2img_width") -- cgit v1.2.3 From 16b9661d2741b241c3964fcbd56559c078b84822 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sun, 1 Jan 2023 09:51:37 +0300 Subject: change karras scheduler sigmas to values recommended by SD from old 0.1 to 10 with an option to revert to old --- modules/sd_samplers.py | 4 +++- modules/shared.py | 6 +++++- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/sd_samplers.py b/modules/sd_samplers.py index 177b5338..e904d860 100644 --- a/modules/sd_samplers.py +++ b/modules/sd_samplers.py @@ -465,7 +465,9 @@ class KDiffusionSampler: if p.sampler_noise_scheduler_override: sigmas = p.sampler_noise_scheduler_override(steps) elif self.config is not None and self.config.options.get('scheduler', None) == 'karras': - sigmas = k_diffusion.sampling.get_sigmas_karras(n=steps, sigma_min=0.1, sigma_max=10, device=shared.device) + sigma_min, sigma_max = (0.1, 10) if opts.use_old_karras_scheduler_sigmas else (self.model_wrap.sigmas[0].item(), self.model_wrap.sigmas[-1].item()) + + sigmas = k_diffusion.sampling.get_sigmas_karras(n=steps, sigma_min=sigma_min, sigma_max=sigma_max, device=shared.device) else: sigmas = self.model_wrap.get_sigmas(steps) diff --git a/modules/shared.py b/modules/shared.py index 948b9542..7f430b93 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -368,13 +368,17 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "img2img_background_color": OptionInfo("#ffffff", "With img2img, fill image's transparent parts with this color.", gr.ColorPicker, {}), "enable_quantization": OptionInfo(False, "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply."), "enable_emphasis": OptionInfo(True, "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention"), - "use_old_emphasis_implementation": OptionInfo(False, "Use old emphasis implementation. Can be useful to reproduce old seeds."), "enable_batch_seeds": OptionInfo(True, "Make K-diffusion samplers produce same images in a batch as when making a single image"), "comma_padding_backtrack": OptionInfo(20, "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens", gr.Slider, {"minimum": 0, "maximum": 74, "step": 1 }), 'CLIP_stop_at_last_layers': OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}), "random_artist_categories": OptionInfo([], "Allowed categories for random artists selection when using the Roll button", gr.CheckboxGroup, {"choices": artist_db.categories()}), })) +options_templates.update(options_section(('compatibility', "Compatibility"), { + "use_old_emphasis_implementation": OptionInfo(False, "Use old emphasis implementation. Can be useful to reproduce old seeds."), + "use_old_karras_scheduler_sigmas": OptionInfo(False, "Use old karras scheduler sigmas (0.1 to 10)."), +})) + options_templates.update(options_section(('interrogate', "Interrogate Options"), { "interrogate_keep_models_in_memory": OptionInfo(False, "Interrogate: keep models in VRAM"), "interrogate_use_builtin_artists": OptionInfo(True, "Interrogate: use artists from artists.csv"), -- cgit v1.2.3 From ef27a18b6b7cb1a8eebdc9b2e88d25baf2c2414d Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Mon, 2 Jan 2023 19:42:10 +0300 Subject: Hires fix rework --- modules/generation_parameters_copypaste.py | 32 ++++++++++++++ modules/images.py | 24 +++++++++-- modules/processing.py | 68 ++++++++++++------------------ modules/shared.py | 7 ++- modules/txt2img.py | 6 +-- modules/ui.py | 15 +++---- scripts/xy_grid.py | 4 +- 7 files changed, 96 insertions(+), 60 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index 8e7f0df0..d6fa822b 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -1,5 +1,6 @@ import base64 import io +import math import os import re from pathlib import Path @@ -164,6 +165,35 @@ def find_hypernetwork_key(hypernet_name, hypernet_hash=None): return None +def restore_old_hires_fix_params(res): + """for infotexts that specify old First pass size parameter, convert it into + width, height, and hr scale""" + + firstpass_width = res.get('First pass size-1', None) + firstpass_height = res.get('First pass size-2', None) + + if firstpass_width is None or firstpass_height is None: + return + + firstpass_width, firstpass_height = int(firstpass_width), int(firstpass_height) + width = int(res.get("Size-1", 512)) + height = int(res.get("Size-2", 512)) + + if firstpass_width == 0 or firstpass_height == 0: + # old algorithm for auto-calculating first pass size + desired_pixel_count = 512 * 512 + actual_pixel_count = width * height + scale = math.sqrt(desired_pixel_count / actual_pixel_count) + firstpass_width = math.ceil(scale * width / 64) * 64 + firstpass_height = math.ceil(scale * height / 64) * 64 + + hr_scale = width / firstpass_width if firstpass_width > 0 else height / firstpass_height + + res['Size-1'] = firstpass_width + res['Size-2'] = firstpass_height + res['Hires upscale'] = hr_scale + + def parse_generation_parameters(x: str): """parses generation parameters string, the one you see in text field under the picture in UI: ``` @@ -221,6 +251,8 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model hypernet_hash = res.get("Hypernet hash", None) res["Hypernet"] = find_hypernetwork_key(hypernet_name, hypernet_hash) + restore_old_hires_fix_params(res) + return res diff --git a/modules/images.py b/modules/images.py index f84fd485..c3a5fc8b 100644 --- a/modules/images.py +++ b/modules/images.py @@ -230,16 +230,32 @@ def draw_prompt_matrix(im, width, height, all_prompts): return draw_grid_annotations(im, width, height, hor_texts, ver_texts) -def resize_image(resize_mode, im, width, height): +def resize_image(resize_mode, im, width, height, upscaler_name=None): + """ + Resizes an image with the specified resize_mode, width, and height. + + Args: + resize_mode: The mode to use when resizing the image. + 0: Resize the image to the specified width and height. + 1: Resize the image to fill the specified width and height, maintaining the aspect ratio, and then center the image within the dimensions, cropping the excess. + 2: Resize the image to fit within the specified width and height, maintaining the aspect ratio, and then center the image within the dimensions, filling empty with data from image. + im: The image to resize. + width: The width to resize the image to. + height: The height to resize the image to. + upscaler_name: The name of the upscaler to use. If not provided, defaults to opts.upscaler_for_img2img. + """ + + upscaler_name = upscaler_name or opts.upscaler_for_img2img + def resize(im, w, h): - if opts.upscaler_for_img2img is None or opts.upscaler_for_img2img == "None" or im.mode == 'L': + if upscaler_name is None or upscaler_name == "None" or im.mode == 'L': return im.resize((w, h), resample=LANCZOS) scale = max(w / im.width, h / im.height) if scale > 1.0: - upscalers = [x for x in shared.sd_upscalers if x.name == opts.upscaler_for_img2img] - assert len(upscalers) > 0, f"could not find upscaler named {opts.upscaler_for_img2img}" + upscalers = [x for x in shared.sd_upscalers if x.name == upscaler_name] + assert len(upscalers) > 0, f"could not find upscaler named {upscaler_name}" upscaler = upscalers[0] im = upscaler.scaler.upscale(im, scale, upscaler.data_path) diff --git a/modules/processing.py b/modules/processing.py index 42dc19ea..4654570c 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -658,14 +658,18 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): sampler = None - def __init__(self, enable_hr: bool=False, denoising_strength: float=0.75, firstphase_width: int=0, firstphase_height: int=0, **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, **kwargs): super().__init__(**kwargs) self.enable_hr = enable_hr self.denoising_strength = denoising_strength - self.firstphase_width = firstphase_width - self.firstphase_height = firstphase_height - self.truncate_x = 0 - self.truncate_y = 0 + self.hr_scale = hr_scale + self.hr_upscaler = hr_upscaler + + if firstphase_width != 0 or firstphase_height != 0: + print("firstphase_width/firstphase_height no longer supported; use hr_scale", file=sys.stderr) + self.hr_scale = self.width / firstphase_width + self.width = firstphase_width + self.height = firstphase_height def init(self, all_prompts, all_seeds, all_subseeds): if self.enable_hr: @@ -674,47 +678,29 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): else: state.job_count = state.job_count * 2 - self.extra_generation_params["First pass size"] = f"{self.firstphase_width}x{self.firstphase_height}" - - if self.firstphase_width == 0 or self.firstphase_height == 0: - desired_pixel_count = 512 * 512 - actual_pixel_count = self.width * self.height - scale = math.sqrt(desired_pixel_count / actual_pixel_count) - self.firstphase_width = math.ceil(scale * self.width / 64) * 64 - self.firstphase_height = math.ceil(scale * self.height / 64) * 64 - firstphase_width_truncated = int(scale * self.width) - firstphase_height_truncated = int(scale * self.height) - - else: - - width_ratio = self.width / self.firstphase_width - height_ratio = self.height / self.firstphase_height - - if width_ratio > height_ratio: - firstphase_width_truncated = self.firstphase_width - firstphase_height_truncated = self.firstphase_width * self.height / self.width - else: - firstphase_width_truncated = self.firstphase_height * self.width / self.height - firstphase_height_truncated = self.firstphase_height - - self.truncate_x = int(self.firstphase_width - firstphase_width_truncated) // opt_f - self.truncate_y = int(self.firstphase_height - firstphase_height_truncated) // opt_f + self.extra_generation_params["Hires upscale"] = self.hr_scale + if self.hr_upscaler is not None: + self.extra_generation_params["Hires upscaler"] = self.hr_upscaler 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_default_mode + if self.enable_hr and latent_scale_mode is None: + assert len([x for x in shared.sd_upscalers if x.name == self.hr_upscaler]) > 0, 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: - 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)) return samples - x = create_random_tensors([opt_C, self.firstphase_height // opt_f, self.firstphase_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, self.firstphase_width, self.firstphase_height)) - - samples = samples[:, :, self.truncate_y//2:samples.shape[2]-self.truncate_y//2, self.truncate_x//2:samples.shape[3]-self.truncate_x//2] + target_width = int(self.width * self.hr_scale) + target_height = int(self.height * self.hr_scale) - """saves image before applying hires fix, if enabled in options; takes as an argument either an image or batch with latent space images""" 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: return @@ -723,11 +709,11 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): images.save_image(image, self.outpath_samples, "", seeds[index], prompts[index], opts.samples_format, suffix="-before-highres-fix") - if opts.use_scale_latent_for_hires_fix: + if latent_scale_mode is not None: for i in range(samples.shape[0]): save_intermediate(samples, i) - samples = torch.nn.functional.interpolate(samples, size=(self.height // opt_f, self.width // opt_f), mode="bilinear") + samples = torch.nn.functional.interpolate(samples, size=(target_height // opt_f, target_width // opt_f), mode=latent_scale_mode) # Avoid making the inpainting conditioning unless necessary as # this does need some extra compute to decode / encode the image again. @@ -747,7 +733,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): save_intermediate(image, i) - image = images.resize_image(0, image, self.width, self.height) + image = images.resize_image(0, image, target_width, target_height, upscaler_name=self.hr_upscaler) image = np.array(image).astype(np.float32) / 255.0 image = np.moveaxis(image, 2, 0) batch_images.append(image) @@ -764,7 +750,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) - noise = create_random_tensors(samples.shape[1:], seeds=seeds, subseeds=subseeds, subseed_strength=subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self) + 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 diff --git a/modules/shared.py b/modules/shared.py index 7f430b93..b65559ee 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -327,7 +327,6 @@ options_templates.update(options_section(('upscaling', "Upscaling"), { "ESRGAN_tile_overlap": OptionInfo(8, "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.", gr.Slider, {"minimum": 0, "maximum": 48, "step": 1}), "realesrgan_enabled_models": OptionInfo(["R-ESRGAN 4x+", "R-ESRGAN 4x+ Anime6B"], "Select which Real-ESRGAN models to show in the web UI. (Requires restart)", gr.CheckboxGroup, lambda: {"choices": realesrgan_models_names()}), "upscaler_for_img2img": OptionInfo(None, "Upscaler for img2img", gr.Dropdown, lambda: {"choices": [x.name for x in sd_upscalers]}), - "use_scale_latent_for_hires_fix": OptionInfo(False, "Upscale latent space image when doing hires. fix"), })) options_templates.update(options_section(('face-restoration', "Face restoration"), { @@ -545,6 +544,12 @@ opts = Options() if os.path.exists(config_filename): opts.load(config_filename) +latent_upscale_default_mode = "Latent" +latent_upscale_modes = { + "Latent": "bilinear", + "Latent (nearest)": "nearest", +} + sd_upscalers = [] sd_model = None diff --git a/modules/txt2img.py b/modules/txt2img.py index 7f61e19a..e189a899 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -8,7 +8,7 @@ import modules.processing as processing from modules.ui import plaintext_to_html -def txt2img(prompt: str, negative_prompt: str, prompt_style: str, prompt_style2: str, 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, firstphase_width: int, firstphase_height: int, *args): +def txt2img(prompt: str, negative_prompt: str, prompt_style: str, prompt_style2: str, 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, *args): p = StableDiffusionProcessingTxt2Img( sd_model=shared.sd_model, outpath_samples=opts.outdir_samples or opts.outdir_txt2img_samples, @@ -33,8 +33,8 @@ def txt2img(prompt: str, negative_prompt: str, prompt_style: str, prompt_style2: tiling=tiling, enable_hr=enable_hr, denoising_strength=denoising_strength if enable_hr else None, - firstphase_width=firstphase_width if enable_hr else None, - firstphase_height=firstphase_height if enable_hr else None, + hr_scale=hr_scale, + hr_upscaler=hr_upscaler, ) p.scripts = modules.scripts.scripts_txt2img diff --git a/modules/ui.py b/modules/ui.py index 7070ea15..27cd9ddd 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -684,11 +684,11 @@ def create_ui(): with gr.Row(): 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='Highres. fix', value=False, elem_id="txt2img_enable_hr") + enable_hr = gr.Checkbox(label='Hires. fix', value=False, elem_id="txt2img_enable_hr") with gr.Row(visible=False) as hr_options: - firstphase_width = gr.Slider(minimum=0, maximum=1024, step=8, label="Firstpass width", value=0, elem_id="txt2img_firstphase_width") - firstphase_height = gr.Slider(minimum=0, maximum=1024, step=8, label="Firstpass height", value=0, elem_id="txt2img_firstphase_height") + 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_scale = gr.Slider(minimum=1.0, maximum=4.0, step=0.05, label="Upscale by", value=2.0, elem_id="txt2img_hr_scale") 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 gr.Row(equal_height=True): @@ -729,8 +729,8 @@ def create_ui(): width, enable_hr, denoising_strength, - firstphase_width, - firstphase_height, + hr_scale, + hr_upscaler, ] + custom_inputs, outputs=[ @@ -762,7 +762,6 @@ def create_ui(): outputs=[hr_options], ) - txt2img_paste_fields = [ (txt2img_prompt, "Prompt"), (txt2img_negative_prompt, "Negative prompt"), @@ -781,8 +780,8 @@ def create_ui(): (denoising_strength, "Denoising strength"), (enable_hr, lambda d: "Denoising strength" in d), (hr_options, lambda d: gr.Row.update(visible="Denoising strength" in d)), - (firstphase_width, "First pass size-1"), - (firstphase_height, "First pass size-2"), + (hr_scale, "Hires upscale"), + (hr_upscaler, "Hires upscaler"), *modules.scripts.scripts_txt2img.infotext_fields ] parameters_copypaste.add_paste_fields("txt2img", None, txt2img_paste_fields) diff --git a/scripts/xy_grid.py b/scripts/xy_grid.py index 3e0b2805..f92f9776 100644 --- a/scripts/xy_grid.py +++ b/scripts/xy_grid.py @@ -202,7 +202,7 @@ axis_options = [ AxisOption("Eta", float, apply_field("eta"), format_value_add_label, None), AxisOption("Clip skip", int, apply_clip_skip, format_value_add_label, None), AxisOption("Denoising", float, apply_field("denoising_strength"), format_value_add_label, None), - AxisOption("Upscale latent space for hires.", str, apply_upscale_latent_space, format_value_add_label, None), + AxisOption("Hires upscaler", str, apply_field("hr_upscaler"), format_value_add_label, None), AxisOption("Cond. Image Mask Weight", float, apply_field("inpainting_mask_weight"), format_value_add_label, None), AxisOption("VAE", str, apply_vae, format_value_add_label, None), AxisOption("Styles", str, apply_styles, format_value_add_label, None), @@ -267,7 +267,6 @@ class SharedSettingsStackHelper(object): self.CLIP_stop_at_last_layers = opts.CLIP_stop_at_last_layers self.hypernetwork = opts.sd_hypernetwork self.model = shared.sd_model - self.use_scale_latent_for_hires_fix = opts.use_scale_latent_for_hires_fix self.vae = opts.sd_vae def __exit__(self, exc_type, exc_value, tb): @@ -278,7 +277,6 @@ class SharedSettingsStackHelper(object): hypernetwork.apply_strength() opts.data["CLIP_stop_at_last_layers"] = self.CLIP_stop_at_last_layers - opts.data["use_scale_latent_for_hires_fix"] = self.use_scale_latent_for_hires_fix re_range = re.compile(r"\s*([+-]?\s*\d+)\s*-\s*([+-]?\s*\d+)(?:\s*\(([+-]\d+)\s*\))?\s*") -- cgit v1.2.3 From a1cf55a9d1c82f8e56c00d549bca5c8fa069f412 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Tue, 3 Jan 2023 10:39:21 +0300 Subject: add option to reorder items in main UI --- modules/shared.py | 13 ++++++ modules/ui.py | 130 +++++++++++++++++++++++++++++++++++------------------- 2 files changed, 97 insertions(+), 46 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index b65559ee..23657a93 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -109,6 +109,17 @@ restricted_opts = { "outdir_save", } +ui_reorder_categories = [ + "sampler", + "dimensions", + "cfg", + "seed", + "checkboxes", + "hires_fix", + "batch", + "scripts", +] + cmd_opts.disable_extension_access = (cmd_opts.share or cmd_opts.listen or cmd_opts.server_name) and not cmd_opts.enable_insecure_extension_access devices.device, devices.device_interrogate, devices.device_gfpgan, devices.device_esrgan, devices.device_codeformer = \ @@ -410,7 +421,9 @@ options_templates.update(options_section(('ui', "User interface"), { "js_modal_lightbox_initially_zoomed": OptionInfo(True, "Show images zoomed in by default in full page image viewer"), "show_progress_in_title": OptionInfo(True, "Show generation progress in window title."), "samplers_in_dropdown": OptionInfo(True, "Use dropdown for sampler selection instead of radio group"), + "dimensions_and_batch_together": OptionInfo(True, "Show Witdth/Height and Batch sliders in same row"), 'quicksettings': OptionInfo("sd_model_checkpoint", "Quicksettings list"), + 'ui_reorder': OptionInfo(", ".join(ui_reorder_categories), "txt2img/ing2img UI item order"), 'localization': OptionInfo("None", "Localization (requires restart)", gr.Dropdown, lambda: {"choices": ["None"] + list(localization.localizations.keys())}, refresh=lambda: localization.list_localizations(cmd_opts.localizations_dir)), })) diff --git a/modules/ui.py b/modules/ui.py index 2c92c422..f2e7c0d6 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -644,6 +644,13 @@ def create_sampler_and_steps_selection(choices, tabname): return steps, sampler_index +def ordered_ui_categories(): + user_order = {x.strip(): i for i, x in enumerate(shared.opts.ui_reorder.split(","))} + + for i, category in sorted(enumerate(shared.ui_reorder_categories), key=lambda x: user_order.get(x[1], x[0] + 1000)): + yield category + + def create_ui(): import modules.img2img import modules.txt2img @@ -672,32 +679,48 @@ def create_ui(): with gr.Row().style(equal_height=False): with gr.Column(variant='panel', elem_id="txt2img_settings"): - steps, sampler_index = create_sampler_and_steps_selection(samplers, "txt2img") - - with FormRow(): - with gr.Column(elem_id="txt2img_column_size", scale=4): - width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="txt2img_width") - height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="txt2img_height") - with gr.Column(elem_id="txt2img_column_batch"): - batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="txt2img_batch_count") - batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="txt2img_batch_size") - - cfg_scale = gr.Slider(minimum=1.0, maximum=30.0, step=0.5, label='CFG Scale', value=7.0, elem_id="txt2img_cfg_scale") - - seed, reuse_seed, subseed, reuse_subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w, seed_checkbox = create_seed_inputs('txt2img') - - with FormRow(elem_id="txt2img_checkboxes"): - 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") + for category in ordered_ui_categories(): + if category == "sampler": + steps, sampler_index = create_sampler_and_steps_selection(samplers, "txt2img") - with FormRow(visible=False) as hr_options: - 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_scale = gr.Slider(minimum=1.0, maximum=4.0, step=0.05, label="Upscale by", value=2.0, elem_id="txt2img_hr_scale") - 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 FormGroup(elem_id="txt2img_script_container"): - custom_inputs = modules.scripts.scripts_txt2img.setup_ui() + elif category == "dimensions": + with FormRow(): + with gr.Column(elem_id="txt2img_column_size", scale=4): + width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="txt2img_width") + height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="txt2img_height") + + if opts.dimensions_and_batch_together: + with gr.Column(elem_id="txt2img_column_batch"): + batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="txt2img_batch_count") + batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="txt2img_batch_size") + + elif category == "cfg": + cfg_scale = gr.Slider(minimum=1.0, maximum=30.0, step=0.5, label='CFG Scale', value=7.0, elem_id="txt2img_cfg_scale") + + elif category == "seed": + seed, reuse_seed, subseed, reuse_subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w, seed_checkbox = create_seed_inputs('txt2img') + + elif category == "checkboxes": + with FormRow(elem_id="txt2img_checkboxes"): + 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") + + elif category == "hires_fix": + with FormRow(visible=False, elem_id="txt2img_hires_fix") as hr_options: + 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_scale = gr.Slider(minimum=1.0, maximum=4.0, step=0.05, label="Upscale by", value=2.0, elem_id="txt2img_hr_scale") + denoising_strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label='Denoising strength', value=0.7, elem_id="txt2img_denoising_strength") + + elif category == "batch": + if not opts.dimensions_and_batch_together: + with FormRow(elem_id="txt2img_column_batch"): + batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="txt2img_batch_count") + batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="txt2img_batch_size") + + elif category == "scripts": + with FormGroup(elem_id="txt2img_script_container"): + custom_inputs = modules.scripts.scripts_txt2img.setup_ui() txt2img_gallery, generation_info, html_info, html_log = create_output_panel("txt2img", opts.outdir_txt2img_samples) parameters_copypaste.bind_buttons({"txt2img": txt2img_paste}, None, txt2img_prompt) @@ -865,28 +888,43 @@ def create_ui(): with FormRow(): resize_mode = gr.Radio(label="Resize mode", elem_id="resize_mode", choices=["Just resize", "Crop and resize", "Resize and fill", "Just resize (latent upscale)"], type="index", value="Just resize") - steps, sampler_index = create_sampler_and_steps_selection(samplers_for_img2img, "img2img") - - with FormRow(): - with gr.Column(elem_id="img2img_column_size", scale=4): - width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="img2img_width") - height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="img2img_height") - with gr.Column(elem_id="img2img_column_batch"): - batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="img2img_batch_count") - batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="img2img_batch_size") - - with FormGroup(): - cfg_scale = gr.Slider(minimum=1.0, maximum=30.0, step=0.5, label='CFG Scale', value=7.0, elem_id="img2img_cfg_scale") - denoising_strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label='Denoising strength', value=0.75, elem_id="img2img_denoising_strength") - - seed, reuse_seed, subseed, reuse_subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w, seed_checkbox = create_seed_inputs('img2img') + for category in ordered_ui_categories(): + if category == "sampler": + steps, sampler_index = create_sampler_and_steps_selection(samplers_for_img2img, "img2img") - with FormRow(elem_id="img2img_checkboxes"): - 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") - - with FormGroup(elem_id="img2img_script_container"): - custom_inputs = modules.scripts.scripts_img2img.setup_ui() + elif category == "dimensions": + with FormRow(): + with gr.Column(elem_id="img2img_column_size", scale=4): + width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="img2img_width") + height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="img2img_height") + + if opts.dimensions_and_batch_together: + with gr.Column(elem_id="img2img_column_batch"): + batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="img2img_batch_count") + batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="img2img_batch_size") + + elif category == "cfg": + with FormGroup(): + cfg_scale = gr.Slider(minimum=1.0, maximum=30.0, step=0.5, label='CFG Scale', value=7.0, elem_id="img2img_cfg_scale") + denoising_strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label='Denoising strength', value=0.75, elem_id="img2img_denoising_strength") + + elif category == "seed": + seed, reuse_seed, subseed, reuse_subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w, seed_checkbox = create_seed_inputs('img2img') + + elif category == "checkboxes": + with FormRow(elem_id="img2img_checkboxes"): + 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") + + elif category == "batch": + if not opts.dimensions_and_batch_together: + with FormRow(elem_id="img2img_column_batch"): + batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="img2img_batch_count") + batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="img2img_batch_size") + + elif category == "scripts": + with FormGroup(elem_id="img2img_script_container"): + custom_inputs = modules.scripts.scripts_img2img.setup_ui() img2img_gallery, generation_info, html_info, html_log = create_output_panel("img2img", opts.outdir_img2img_samples) parameters_copypaste.bind_buttons({"img2img": img2img_paste}, None, img2img_prompt) -- cgit v1.2.3 From bddebe09edeb6a18f2c06986d5658a7be3a563ea Mon Sep 17 00:00:00 2001 From: Shondoit Date: Tue, 3 Jan 2023 10:26:37 +0100 Subject: Save Optimizer next to TI embedding Also add check to load only .PT and .BIN files as embeddings. (since we add .optim files in the same directory) --- modules/shared.py | 2 +- modules/textual_inversion/textual_inversion.py | 40 ++++++++++++++++++++------ 2 files changed, 33 insertions(+), 9 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index 23657a93..c541d18c 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -355,7 +355,7 @@ options_templates.update(options_section(('system', "System"), { options_templates.update(options_section(('training', "Training"), { "unload_models_when_training": OptionInfo(False, "Move VAE and CLIP to RAM when training if possible. Saves VRAM."), "pin_memory": OptionInfo(False, "Turn on pin_memory for DataLoader. Makes training slightly faster but can increase memory usage."), - "save_optimizer_state": OptionInfo(False, "Saves Optimizer state as separate *.optim file. Training can be resumed with HN itself and matching optim file."), + "save_optimizer_state": OptionInfo(False, "Saves Optimizer state as separate *.optim file. Training of embedding or HN can be resumed with the matching optim file."), "dataset_filename_word_regex": OptionInfo("", "Filename word regex"), "dataset_filename_join_string": OptionInfo(" ", "Filename join string"), "training_image_repeats_per_epoch": OptionInfo(1, "Number of repeats for a single input image per epoch; used only for displaying epoch number", gr.Number, {"precision": 0}), diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index fd253477..16176e90 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -28,6 +28,7 @@ class Embedding: self.cached_checksum = None self.sd_checkpoint = None self.sd_checkpoint_name = None + self.optimizer_state_dict = None def save(self, filename): embedding_data = { @@ -41,6 +42,13 @@ class Embedding: torch.save(embedding_data, filename) + if shared.opts.save_optimizer_state and self.optimizer_state_dict is not None: + optimizer_saved_dict = { + 'hash': self.checksum(), + 'optimizer_state_dict': self.optimizer_state_dict, + } + torch.save(optimizer_saved_dict, filename + '.optim') + def checksum(self): if self.cached_checksum is not None: return self.cached_checksum @@ -95,9 +103,10 @@ class EmbeddingDatabase: self.expected_shape = self.get_expected_shape() def process_file(path, filename): - name = os.path.splitext(filename)[0] + name, ext = os.path.splitext(filename) + ext = ext.upper() - if os.path.splitext(filename.upper())[-1] in ['.PNG', '.WEBP', '.JXL', '.AVIF']: + if ext in ['.PNG', '.WEBP', '.JXL', '.AVIF']: embed_image = Image.open(path) if hasattr(embed_image, 'text') and 'sd-ti-embedding' in embed_image.text: data = embedding_from_b64(embed_image.text['sd-ti-embedding']) @@ -105,8 +114,10 @@ class EmbeddingDatabase: else: data = extract_image_data_embed(embed_image) name = data.get('name', name) - else: + elif ext in ['.BIN', '.PT']: data = torch.load(path, map_location="cpu") + else: + return # textual inversion embeddings if 'string_to_param' in data: @@ -300,6 +311,20 @@ def train_embedding(embedding_name, learn_rate, batch_size, gradient_step, data_ embedding.vec.requires_grad = True optimizer = torch.optim.AdamW([embedding.vec], lr=scheduler.learn_rate, weight_decay=0.0) + if shared.opts.save_optimizer_state: + optimizer_state_dict = None + if os.path.exists(filename + '.optim'): + optimizer_saved_dict = torch.load(filename + '.optim', map_location='cpu') + if embedding.checksum() == optimizer_saved_dict.get('hash', None): + optimizer_state_dict = optimizer_saved_dict.get('optimizer_state_dict', None) + + if optimizer_state_dict is not None: + optimizer.load_state_dict(optimizer_state_dict) + print("Loaded existing optimizer from checkpoint") + else: + print("No saved optimizer exists in checkpoint") + + scaler = torch.cuda.amp.GradScaler() batch_size = ds.batch_size @@ -366,9 +391,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, gradient_step, data_ # Before saving, change name to match current checkpoint. embedding_name_every = f'{embedding_name}-{steps_done}' last_saved_file = os.path.join(embedding_dir, f'{embedding_name_every}.pt') - #if shared.opts.save_optimizer_state: - #embedding.optimizer_state_dict = optimizer.state_dict() - save_embedding(embedding, checkpoint, embedding_name_every, last_saved_file, remove_cached_checksum=True) + save_embedding(embedding, optimizer, checkpoint, embedding_name_every, last_saved_file, remove_cached_checksum=True) embedding_yet_to_be_embedded = True write_loss(log_directory, "textual_inversion_loss.csv", embedding.step, steps_per_epoch, { @@ -458,7 +481,7 @@ Last saved image: {html.escape(last_saved_image)}

""" filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt') - save_embedding(embedding, checkpoint, embedding_name, filename, remove_cached_checksum=True) + save_embedding(embedding, optimizer, checkpoint, embedding_name, filename, remove_cached_checksum=True) except Exception: print(traceback.format_exc(), file=sys.stderr) pass @@ -470,7 +493,7 @@ Last saved image: {html.escape(last_saved_image)}
return embedding, filename -def save_embedding(embedding, checkpoint, embedding_name, filename, remove_cached_checksum=True): +def save_embedding(embedding, optimizer, checkpoint, embedding_name, filename, remove_cached_checksum=True): old_embedding_name = embedding.name old_sd_checkpoint = embedding.sd_checkpoint if hasattr(embedding, "sd_checkpoint") else None old_sd_checkpoint_name = embedding.sd_checkpoint_name if hasattr(embedding, "sd_checkpoint_name") else None @@ -481,6 +504,7 @@ def save_embedding(embedding, checkpoint, embedding_name, filename, remove_cache if remove_cached_checksum: embedding.cached_checksum = None embedding.name = embedding_name + embedding.optimizer_state_dict = optimizer.state_dict() embedding.save(filename) except: embedding.sd_checkpoint = old_sd_checkpoint -- cgit v1.2.3 From aaa4c2aacbb6523077334093c81bd475d757f7a1 Mon Sep 17 00:00:00 2001 From: Vladimir Mandic Date: Tue, 3 Jan 2023 09:45:16 -0500 Subject: add api logging --- modules/api/api.py | 24 +++++++++++++++++++++++- modules/shared.py | 1 + 2 files changed, 24 insertions(+), 1 deletion(-) (limited to 'modules/shared.py') diff --git a/modules/api/api.py b/modules/api/api.py index 9c670f00..53135470 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -1,11 +1,12 @@ import base64 import io import time +import datetime import uvicorn from threading import Lock from io import BytesIO from gradio.processing_utils import decode_base64_to_file -from fastapi import APIRouter, Depends, FastAPI, HTTPException +from fastapi import APIRouter, Depends, FastAPI, HTTPException, Request, Response from fastapi.security import HTTPBasic, HTTPBasicCredentials from secrets import compare_digest @@ -67,6 +68,26 @@ def encode_pil_to_base64(image): bytes_data = output_bytes.getvalue() return base64.b64encode(bytes_data) +def init_api_middleware(app: FastAPI): + @app.middleware("http") + async def log_and_time(req: Request, call_next): + ts = time.time() + res: Response = await call_next(req) + duration = str(round(time.time() - ts, 4)) + res.headers["X-Process-Time"] = duration + if shared.cmd_opts.api_log: + print('API {t} {code} {prot}/{ver} {method} {p} {cli} {duration}'.format( + t = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S.%f"), + code = res.status_code, + ver = req.scope.get('http_version', '0.0'), + cli = req.scope.get('client', ('0:0.0.0', 0))[0], + prot = req.scope.get('scheme', 'err'), + method = req.scope.get('method', 'err'), + p = req.scope.get('path', 'err'), + duration = duration, + )) + return res + class Api: def __init__(self, app: FastAPI, queue_lock: Lock): @@ -78,6 +99,7 @@ class Api: self.router = APIRouter() self.app = app + init_api_middleware(self.app) self.queue_lock = queue_lock self.add_api_route("/sdapi/v1/txt2img", self.text2imgapi, methods=["POST"], response_model=TextToImageResponse) self.add_api_route("/sdapi/v1/img2img", self.img2imgapi, methods=["POST"], response_model=ImageToImageResponse) diff --git a/modules/shared.py b/modules/shared.py index 23657a93..2a03d716 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -82,6 +82,7 @@ parser.add_argument('--vae-path', type=str, help='Path to Variational Autoencode parser.add_argument("--disable-safe-unpickle", action='store_true', help="disable checking pytorch models for malicious code", default=False) parser.add_argument("--api", action='store_true', help="use api=True to launch the API together with the webui (use --nowebui instead for only the API)") parser.add_argument("--api-auth", type=str, help='Set authentication for API like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"', default=None) +parser.add_argument("--api-log", action='store_true', help="use api-log=True to enable logging of all API requests") parser.add_argument("--nowebui", action='store_true', help="use api=True to launch the API instead of the webui") parser.add_argument("--ui-debug-mode", action='store_true', help="Don't load model to quickly launch UI") parser.add_argument("--device-id", type=str, help="Select the default CUDA device to use (export CUDA_VISIBLE_DEVICES=0,1,etc might be needed before)", default=None) -- cgit v1.2.3 From 02d7abf5141431b9a3a8a189bb3136c71abd5e79 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Wed, 4 Jan 2023 12:35:07 +0300 Subject: helpful error message when trying to load 2.0 without config failing to load model weights from settings won't break generation for currently loaded model anymore --- modules/errors.py | 25 +++++++++++++++++++++++-- modules/sd_models.py | 26 ++++++++++++++++++-------- modules/shared.py | 9 +++++++-- webui.py | 12 ++++++++++-- 4 files changed, 58 insertions(+), 14 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/errors.py b/modules/errors.py index 372dc51a..a668c014 100644 --- a/modules/errors.py +++ b/modules/errors.py @@ -2,9 +2,30 @@ import sys import traceback +def print_error_explanation(message): + lines = message.strip().split("\n") + max_len = max([len(x) for x in lines]) + + print('=' * max_len, file=sys.stderr) + for line in lines: + print(line, file=sys.stderr) + print('=' * max_len, file=sys.stderr) + + +def display(e: Exception, task): + print(f"{task or 'error'}: {type(e).__name__}", file=sys.stderr) + print(traceback.format_exc(), file=sys.stderr) + + message = str(e) + if "copying a param with shape torch.Size([640, 1024]) from checkpoint, the shape in current model is torch.Size([640, 768])" in message: + print_error_explanation(""" +The most likely cause of this is you are trying to load Stable Diffusion 2.0 model without specifying its connfig file. +See https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#stable-diffusion-20 for how to solve this. + """) + + def run(code, task): try: code() except Exception as e: - print(f"{task}: {type(e).__name__}", file=sys.stderr) - print(traceback.format_exc(), file=sys.stderr) + display(task, e) diff --git a/modules/sd_models.py b/modules/sd_models.py index b98b05fc..6846b74a 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -278,6 +278,7 @@ def enable_midas_autodownload(): midas.api.load_model = load_model_wrapper + def load_model(checkpoint_info=None): from modules import lowvram, sd_hijack checkpoint_info = checkpoint_info or select_checkpoint() @@ -312,6 +313,7 @@ def load_model(checkpoint_info=None): sd_config.model.params.unet_config.params.use_fp16 = False sd_model = instantiate_from_config(sd_config.model) + load_model_weights(sd_model, checkpoint_info) if shared.cmd_opts.lowvram or shared.cmd_opts.medvram: @@ -336,10 +338,12 @@ def load_model(checkpoint_info=None): def reload_model_weights(sd_model=None, info=None): from modules import lowvram, devices, sd_hijack checkpoint_info = info or select_checkpoint() - + if not sd_model: sd_model = shared.sd_model + current_checkpoint_info = sd_model.sd_checkpoint_info + if sd_model.sd_model_checkpoint == checkpoint_info.filename: return @@ -356,13 +360,19 @@ def reload_model_weights(sd_model=None, info=None): sd_hijack.model_hijack.undo_hijack(sd_model) - load_model_weights(sd_model, checkpoint_info) - - sd_hijack.model_hijack.hijack(sd_model) - script_callbacks.model_loaded_callback(sd_model) - - if not shared.cmd_opts.lowvram and not shared.cmd_opts.medvram: - sd_model.to(devices.device) + try: + load_model_weights(sd_model, checkpoint_info) + except Exception as e: + print("Failed to load checkpoint, restoring previous") + load_model_weights(sd_model, current_checkpoint_info) + raise + finally: + sd_hijack.model_hijack.hijack(sd_model) + script_callbacks.model_loaded_callback(sd_model) + + if not shared.cmd_opts.lowvram and not shared.cmd_opts.medvram: + sd_model.to(devices.device) print("Weights loaded.") + return sd_model diff --git a/modules/shared.py b/modules/shared.py index 23657a93..7588c47b 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -14,7 +14,7 @@ import modules.interrogate import modules.memmon import modules.styles import modules.devices as devices -from modules import localization, sd_vae, extensions, script_loading +from modules import localization, sd_vae, extensions, script_loading, errors from modules.paths import models_path, script_path, sd_path @@ -494,7 +494,12 @@ class Options: return False if self.data_labels[key].onchange is not None: - self.data_labels[key].onchange() + try: + self.data_labels[key].onchange() + except Exception as e: + errors.display(e, f"changing setting {key} to {value}") + setattr(self, key, oldval) + return False return True diff --git a/webui.py b/webui.py index c7d55a97..13375e71 100644 --- a/webui.py +++ b/webui.py @@ -9,7 +9,7 @@ from fastapi import FastAPI from fastapi.middleware.cors import CORSMiddleware from fastapi.middleware.gzip import GZipMiddleware -from modules import import_hook +from modules import import_hook, errors from modules.call_queue import wrap_queued_call, queue_lock, wrap_gradio_gpu_call from modules.paths import script_path @@ -61,7 +61,15 @@ def initialize(): modelloader.load_upscalers() modules.sd_vae.refresh_vae_list() - modules.sd_models.load_model() + + try: + modules.sd_models.load_model() + except Exception as e: + errors.display(e, "loading stable diffusion model") + print("", file=sys.stderr) + print("Stable diffusion model failed to load, exiting", file=sys.stderr) + exit(1) + shared.opts.onchange("sd_model_checkpoint", wrap_queued_call(lambda: modules.sd_models.reload_model_weights())) shared.opts.onchange("sd_vae", wrap_queued_call(lambda: modules.sd_vae.reload_vae_weights()), call=False) shared.opts.onchange("sd_vae_as_default", wrap_queued_call(lambda: modules.sd_vae.reload_vae_weights()), call=False) -- cgit v1.2.3 From 96cf15bedecbed97ef9b70b8413d543a9aee5adf Mon Sep 17 00:00:00 2001 From: MMaker Date: Wed, 4 Jan 2023 05:12:06 -0500 Subject: Add new latent upscale modes --- modules/shared.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index 7588c47b..a10f69a9 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -564,8 +564,11 @@ if os.path.exists(config_filename): latent_upscale_default_mode = "Latent" latent_upscale_modes = { - "Latent": "bilinear", - "Latent (nearest)": "nearest", + "Latent": {"mode": "bilinear", "antialias": False}, + "Latent (antialiased)": {"mode": "bilinear", "antialias": True}, + "Latent (bicubic)": {"mode": "bicubic", "antialias": False}, + "Latent (bicubic, antialiased)": {"mode": "bicubic", "antialias": True}, + "Latent (nearest)": {"mode": "nearest", "antialias": False}, } sd_upscalers = [] -- cgit v1.2.3 From b2151b934fe0a3613570c6abd7615d3788fd1c8f Mon Sep 17 00:00:00 2001 From: MMaker Date: Wed, 4 Jan 2023 05:36:18 -0500 Subject: Rename bicubic antialiased option Comma was causing the the value in PNG info to be quoted, which causes the upscaler dropdown option to be blank when sending to UI --- modules/shared.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index a10f69a9..c1b20081 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -567,7 +567,7 @@ latent_upscale_modes = { "Latent": {"mode": "bilinear", "antialias": False}, "Latent (antialiased)": {"mode": "bilinear", "antialias": True}, "Latent (bicubic)": {"mode": "bicubic", "antialias": False}, - "Latent (bicubic, antialiased)": {"mode": "bicubic", "antialias": True}, + "Latent (bicubic antialiased)": {"mode": "bicubic", "antialias": True}, "Latent (nearest)": {"mode": "nearest", "antialias": False}, } -- cgit v1.2.3 From 1cfd8aec4ae5a6ca1afd67b44cb4ef6dd14d8c34 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Wed, 4 Jan 2023 16:05:42 +0300 Subject: make it possible to work with opts.show_progress_every_n_steps = -1 with medvram --- modules/shared.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/shared.py b/modules/shared.py index 4fcc6edd..54a6ba23 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -214,12 +214,13 @@ class State: """sets self.current_image from self.current_latent if enough sampling steps have been made after the last call to this""" def set_current_image(self): + if not parallel_processing_allowed: + return + if self.sampling_step - self.current_image_sampling_step >= opts.show_progress_every_n_steps and opts.show_progress_every_n_steps > 0: self.do_set_current_image() def do_set_current_image(self): - if not parallel_processing_allowed: - return if self.current_latent is None: return @@ -231,6 +232,7 @@ class State: self.current_image_sampling_step = self.sampling_step + state = State() artist_db = modules.artists.ArtistsDatabase(os.path.join(script_path, 'artists.csv')) -- cgit v1.2.3 From bc43293c640aef65df3136de9e5bd8b7e79eb3e0 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Wed, 4 Jan 2023 23:56:43 +0300 Subject: fix incorrect display/calculation for number of steps for hires fix in progress bars --- modules/processing.py | 9 ++++++--- modules/sd_samplers.py | 5 +++-- modules/shared.py | 4 +++- 3 files changed, 12 insertions(+), 6 deletions(-) (limited to 'modules/shared.py') diff --git a/modules/processing.py b/modules/processing.py index 9cad05f2..f28e7212 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -685,10 +685,13 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def init(self, all_prompts, all_seeds, all_subseeds): if self.enable_hr: - if state.job_count == -1: - state.job_count = self.n_iter * 2 - else: + if not state.processing_has_refined_job_count: + if state.job_count == -1: + state.job_count = self.n_iter + + shared.total_tqdm.updateTotal((self.steps + (self.hr_second_pass_steps or self.steps)) * state.job_count) state.job_count = state.job_count * 2 + state.processing_has_refined_job_count = True if self.hr_resize_x == 0 and self.hr_resize_y == 0: self.extra_generation_params["Hires upscale"] = self.hr_scale diff --git a/modules/sd_samplers.py b/modules/sd_samplers.py index e904d860..3851a77f 100644 --- a/modules/sd_samplers.py +++ b/modules/sd_samplers.py @@ -97,8 +97,9 @@ sampler_extra_params = { def setup_img2img_steps(p, steps=None): if opts.img2img_fix_steps or steps is not None: - steps = int((steps or p.steps) / min(p.denoising_strength, 0.999)) if p.denoising_strength > 0 else 0 - t_enc = p.steps - 1 + requested_steps = (steps or p.steps) + steps = int(requested_steps / min(p.denoising_strength, 0.999)) if p.denoising_strength > 0 else 0 + t_enc = requested_steps - 1 else: steps = p.steps t_enc = int(min(p.denoising_strength, 0.999) * steps) diff --git a/modules/shared.py b/modules/shared.py index 54a6ba23..04c545ee 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -153,6 +153,7 @@ class State: job = "" job_no = 0 job_count = 0 + processing_has_refined_job_count = False job_timestamp = '0' sampling_step = 0 sampling_steps = 0 @@ -194,6 +195,7 @@ class State: def begin(self): self.sampling_step = 0 self.job_count = -1 + self.processing_has_refined_job_count = False self.job_no = 0 self.job_timestamp = datetime.datetime.now().strftime("%Y%m%d%H%M%S") self.current_latent = None @@ -608,7 +610,7 @@ class TotalTQDM: return if self._tqdm is None: self.reset() - self._tqdm.total=new_total + self._tqdm.total = new_total def clear(self): if self._tqdm is not None: -- cgit v1.2.3