From d380f939b5ab6a28bed6d1de3cf283e194255963 Mon Sep 17 00:00:00 2001 From: Leon Feng <523684+leon0707@users.noreply.github.com> Date: Sat, 15 Jul 2023 23:31:59 -0400 Subject: [PATCH 001/147] Update shared.py --- modules/shared.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/shared.py b/modules/shared.py index a0862055..564799bc 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -394,7 +394,7 @@ 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_model_checkpoint": OptionInfo("", "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("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"), From b75b004fe62826455f1aa77e849e7da13902cb17 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 16 Jul 2023 23:13:55 +0300 Subject: [PATCH 002/147] lora extension rework to include other types of networks --- .../Lora/extra_networks_lora.py | 18 +- extensions-builtin/Lora/lora.py | 537 ------------------ extensions-builtin/Lora/lyco_helpers.py | 15 + extensions-builtin/Lora/network.py | 98 ++++ extensions-builtin/Lora/network_hada.py | 59 ++ extensions-builtin/Lora/network_lora.py | 70 +++ extensions-builtin/Lora/network_lyco.py | 39 ++ extensions-builtin/Lora/networks.py | 443 +++++++++++++++ .../Lora/scripts/lora_script.py | 79 +-- .../Lora/ui_extra_networks_lora.py | 8 +- 10 files changed, 777 insertions(+), 589 deletions(-) delete mode 100644 extensions-builtin/Lora/lora.py create mode 100644 extensions-builtin/Lora/lyco_helpers.py create mode 100644 extensions-builtin/Lora/network.py create mode 100644 extensions-builtin/Lora/network_hada.py create mode 100644 extensions-builtin/Lora/network_lora.py create mode 100644 extensions-builtin/Lora/network_lyco.py create mode 100644 extensions-builtin/Lora/networks.py diff --git a/extensions-builtin/Lora/extra_networks_lora.py b/extensions-builtin/Lora/extra_networks_lora.py index 66ee9c85..8a6639cf 100644 --- a/extensions-builtin/Lora/extra_networks_lora.py +++ b/extensions-builtin/Lora/extra_networks_lora.py @@ -1,5 +1,5 @@ from modules import extra_networks, shared -import lora +import networks class ExtraNetworkLora(extra_networks.ExtraNetwork): @@ -9,7 +9,7 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): def activate(self, p, params_list): additional = shared.opts.sd_lora - if additional != "None" and additional in lora.available_loras and not any(x for x in params_list if x.items[0] == additional): + 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])) @@ -21,12 +21,12 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): names.append(params.items[0]) multipliers.append(float(params.items[1]) if len(params.items) > 1 else 1.0) - lora.load_loras(names, multipliers) + networks.load_networks(names, multipliers) if shared.opts.lora_add_hashes_to_infotext: - lora_hashes = [] - for item in lora.loaded_loras: - shorthash = item.lora_on_disk.shorthash + network_hashes = [] + for item in networks.loaded_networks: + shorthash = item.network_on_disk.shorthash if not shorthash: continue @@ -36,10 +36,10 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): alias = alias.replace(":", "").replace(",", "") - lora_hashes.append(f"{alias}: {shorthash}") + network_hashes.append(f"{alias}: {shorthash}") - if lora_hashes: - p.extra_generation_params["Lora hashes"] = ", ".join(lora_hashes) + if network_hashes: + p.extra_generation_params["Lora hashes"] = ", ".join(network_hashes) def deactivate(self, p): pass diff --git a/extensions-builtin/Lora/lora.py b/extensions-builtin/Lora/lora.py deleted file mode 100644 index 9cdff6ed..00000000 --- a/extensions-builtin/Lora/lora.py +++ /dev/null @@ -1,537 +0,0 @@ -import os -import re -import torch -from typing import Union - -from modules import shared, devices, sd_models, errors, scripts, sd_hijack, hashes, cache - -metadata_tags_order = {"ss_sd_model_name": 1, "ss_resolution": 2, "ss_clip_skip": 3, "ss_num_train_images": 10, "ss_tag_frequency": 20} - -re_digits = re.compile(r"\d+") -re_x_proj = re.compile(r"(.*)_([qkv]_proj)$") -re_compiled = {} - -suffix_conversion = { - "attentions": {}, - "resnets": { - "conv1": "in_layers_2", - "conv2": "out_layers_3", - "time_emb_proj": "emb_layers_1", - "conv_shortcut": "skip_connection", - } -} - - -def convert_diffusers_name_to_compvis(key, is_sd2): - def match(match_list, regex_text): - regex = re_compiled.get(regex_text) - if regex is None: - regex = re.compile(regex_text) - re_compiled[regex_text] = regex - - r = re.match(regex, key) - if not r: - return False - - match_list.clear() - match_list.extend([int(x) if re.match(re_digits, x) else x for x in r.groups()]) - return True - - m = [] - - if match(m, r"lora_unet_down_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"): - suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3]) - return f"diffusion_model_input_blocks_{1 + m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}" - - if match(m, r"lora_unet_mid_block_(attentions|resnets)_(\d+)_(.+)"): - suffix = suffix_conversion.get(m[0], {}).get(m[2], m[2]) - return f"diffusion_model_middle_block_{1 if m[0] == 'attentions' else m[1] * 2}_{suffix}" - - if match(m, r"lora_unet_up_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"): - suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3]) - return f"diffusion_model_output_blocks_{m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}" - - if match(m, r"lora_unet_down_blocks_(\d+)_downsamplers_0_conv"): - return f"diffusion_model_input_blocks_{3 + m[0] * 3}_0_op" - - if match(m, r"lora_unet_up_blocks_(\d+)_upsamplers_0_conv"): - return f"diffusion_model_output_blocks_{2 + m[0] * 3}_{2 if m[0]>0 else 1}_conv" - - if match(m, r"lora_te_text_model_encoder_layers_(\d+)_(.+)"): - if is_sd2: - if 'mlp_fc1' in m[1]: - return f"model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc1', 'mlp_c_fc')}" - elif 'mlp_fc2' in m[1]: - return f"model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc2', 'mlp_c_proj')}" - else: - return f"model_transformer_resblocks_{m[0]}_{m[1].replace('self_attn', 'attn')}" - - return f"transformer_text_model_encoder_layers_{m[0]}_{m[1]}" - - if match(m, r"lora_te2_text_model_encoder_layers_(\d+)_(.+)"): - if 'mlp_fc1' in m[1]: - return f"1_model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc1', 'mlp_c_fc')}" - elif 'mlp_fc2' in m[1]: - return f"1_model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc2', 'mlp_c_proj')}" - else: - return f"1_model_transformer_resblocks_{m[0]}_{m[1].replace('self_attn', 'attn')}" - - return key - - -class LoraOnDisk: - def __init__(self, name, filename): - self.name = name - self.filename = filename - self.metadata = {} - self.is_safetensors = os.path.splitext(filename)[1].lower() == ".safetensors" - - def read_metadata(): - metadata = sd_models.read_metadata_from_safetensors(filename) - metadata.pop('ssmd_cover_images', None) # those are cover images, and they are too big to display in UI as text - - return metadata - - if self.is_safetensors: - try: - self.metadata = cache.cached_data_for_file('safetensors-metadata', "lora/" + self.name, filename, read_metadata) - except Exception as e: - errors.display(e, f"reading lora {filename}") - - if self.metadata: - m = {} - for k, v in sorted(self.metadata.items(), key=lambda x: metadata_tags_order.get(x[0], 999)): - m[k] = v - - self.metadata = m - - self.alias = self.metadata.get('ss_output_name', self.name) - - self.hash = None - self.shorthash = None - self.set_hash( - self.metadata.get('sshs_model_hash') or - hashes.sha256_from_cache(self.filename, "lora/" + self.name, use_addnet_hash=self.is_safetensors) or - '' - ) - - def set_hash(self, v): - self.hash = v - self.shorthash = self.hash[0:12] - - if self.shorthash: - available_lora_hash_lookup[self.shorthash] = self - - def read_hash(self): - if not self.hash: - self.set_hash(hashes.sha256(self.filename, "lora/" + self.name, use_addnet_hash=self.is_safetensors) or '') - - def get_alias(self): - if shared.opts.lora_preferred_name == "Filename" or self.alias.lower() in forbidden_lora_aliases: - return self.name - else: - return self.alias - - -class LoraModule: - def __init__(self, name, lora_on_disk: LoraOnDisk): - self.name = name - self.lora_on_disk = lora_on_disk - self.multiplier = 1.0 - self.modules = {} - self.mtime = None - - self.mentioned_name = None - """the text that was used to add lora to prompt - can be either name or an alias""" - - -class LoraUpDownModule: - def __init__(self): - self.up = None - self.down = None - self.alpha = None - - -def assign_lora_names_to_compvis_modules(sd_model): - lora_layer_mapping = {} - - if shared.sd_model.is_sdxl: - for i, embedder in enumerate(shared.sd_model.conditioner.embedders): - if not hasattr(embedder, 'wrapped'): - continue - - for name, module in embedder.wrapped.named_modules(): - lora_name = f'{i}_{name.replace(".", "_")}' - lora_layer_mapping[lora_name] = module - module.lora_layer_name = lora_name - else: - for name, module in shared.sd_model.cond_stage_model.wrapped.named_modules(): - lora_name = name.replace(".", "_") - lora_layer_mapping[lora_name] = module - module.lora_layer_name = lora_name - - for name, module in shared.sd_model.model.named_modules(): - lora_name = name.replace(".", "_") - lora_layer_mapping[lora_name] = module - module.lora_layer_name = lora_name - - sd_model.lora_layer_mapping = lora_layer_mapping - - -def load_lora(name, lora_on_disk): - lora = LoraModule(name, lora_on_disk) - lora.mtime = os.path.getmtime(lora_on_disk.filename) - - sd = sd_models.read_state_dict(lora_on_disk.filename) - - # this should not be needed but is here as an emergency fix for an unknown error people are experiencing in 1.2.0 - if not hasattr(shared.sd_model, 'lora_layer_mapping'): - assign_lora_names_to_compvis_modules(shared.sd_model) - - keys_failed_to_match = {} - is_sd2 = 'model_transformer_resblocks' in shared.sd_model.lora_layer_mapping - - for key_lora, weight in sd.items(): - key_lora_without_lora_parts, lora_key = key_lora.split(".", 1) - - key = convert_diffusers_name_to_compvis(key_lora_without_lora_parts, is_sd2) - sd_module = shared.sd_model.lora_layer_mapping.get(key, None) - - if sd_module is None: - m = re_x_proj.match(key) - if m: - sd_module = shared.sd_model.lora_layer_mapping.get(m.group(1), None) - - # SDXL loras seem to already have correct compvis keys, so only need to replace "lora_unet" with "diffusion_model" - if sd_module is None and "lora_unet" in key_lora_without_lora_parts: - key = key_lora_without_lora_parts.replace("lora_unet", "diffusion_model") - sd_module = shared.sd_model.lora_layer_mapping.get(key, None) - elif sd_module is None and "lora_te1_text_model" in key_lora_without_lora_parts: - key = key_lora_without_lora_parts.replace("lora_te1_text_model", "0_transformer_text_model") - sd_module = shared.sd_model.lora_layer_mapping.get(key, None) - - if sd_module is None: - keys_failed_to_match[key_lora] = key - continue - - lora_module = lora.modules.get(key, None) - if lora_module is None: - lora_module = LoraUpDownModule() - lora.modules[key] = lora_module - - if lora_key == "alpha": - lora_module.alpha = weight.item() - continue - - if type(sd_module) == torch.nn.Linear: - module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) - elif type(sd_module) == torch.nn.modules.linear.NonDynamicallyQuantizableLinear: - module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) - elif type(sd_module) == torch.nn.MultiheadAttention: - module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) - elif type(sd_module) == torch.nn.Conv2d and weight.shape[2:] == (1, 1): - module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (1, 1), bias=False) - elif type(sd_module) == torch.nn.Conv2d and weight.shape[2:] == (3, 3): - module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (3, 3), bias=False) - else: - print(f'Lora layer {key_lora} matched a layer with unsupported type: {type(sd_module).__name__}') - continue - raise AssertionError(f"Lora layer {key_lora} matched a layer with unsupported type: {type(sd_module).__name__}") - - with torch.no_grad(): - module.weight.copy_(weight) - - module.to(device=devices.cpu, dtype=devices.dtype) - - if lora_key == "lora_up.weight": - lora_module.up = module - elif lora_key == "lora_down.weight": - lora_module.down = module - else: - raise AssertionError(f"Bad Lora layer name: {key_lora} - must end in lora_up.weight, lora_down.weight or alpha") - - if keys_failed_to_match: - print(f"Failed to match keys when loading Lora {lora_on_disk.filename}: {keys_failed_to_match}") - - return lora - - -def load_loras(names, multipliers=None): - already_loaded = {} - - for lora in loaded_loras: - if lora.name in names: - already_loaded[lora.name] = lora - - loaded_loras.clear() - - loras_on_disk = [available_lora_aliases.get(name, None) for name in names] - if any(x is None for x in loras_on_disk): - list_available_loras() - - loras_on_disk = [available_lora_aliases.get(name, None) for name in names] - - failed_to_load_loras = [] - - for i, name in enumerate(names): - lora = already_loaded.get(name, None) - - lora_on_disk = loras_on_disk[i] - - if lora_on_disk is not None: - if lora is None or os.path.getmtime(lora_on_disk.filename) > lora.mtime: - try: - lora = load_lora(name, lora_on_disk) - except Exception as e: - errors.display(e, f"loading Lora {lora_on_disk.filename}") - continue - - lora.mentioned_name = name - - lora_on_disk.read_hash() - - if lora is None: - failed_to_load_loras.append(name) - print(f"Couldn't find Lora with name {name}") - continue - - lora.multiplier = multipliers[i] if multipliers else 1.0 - loaded_loras.append(lora) - - if failed_to_load_loras: - sd_hijack.model_hijack.comments.append("Failed to find Loras: " + ", ".join(failed_to_load_loras)) - - -def lora_calc_updown(lora, module, target): - with torch.no_grad(): - up = module.up.weight.to(target.device, dtype=target.dtype) - down = module.down.weight.to(target.device, dtype=target.dtype) - - if up.shape[2:] == (1, 1) and down.shape[2:] == (1, 1): - updown = (up.squeeze(2).squeeze(2) @ down.squeeze(2).squeeze(2)).unsqueeze(2).unsqueeze(3) - elif up.shape[2:] == (3, 3) or down.shape[2:] == (3, 3): - updown = torch.nn.functional.conv2d(down.permute(1, 0, 2, 3), up).permute(1, 0, 2, 3) - else: - updown = up @ down - - updown = updown * lora.multiplier * (module.alpha / module.up.weight.shape[1] if module.alpha else 1.0) - - return updown - - -def lora_restore_weights_from_backup(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn.MultiheadAttention]): - weights_backup = getattr(self, "lora_weights_backup", None) - - if weights_backup is None: - return - - if isinstance(self, torch.nn.MultiheadAttention): - self.in_proj_weight.copy_(weights_backup[0]) - self.out_proj.weight.copy_(weights_backup[1]) - else: - self.weight.copy_(weights_backup) - - -def lora_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn.MultiheadAttention]): - """ - Applies the currently selected set of Loras to the weights of torch layer self. - If weights already have this particular set of loras applied, does nothing. - If not, restores orginal weights from backup and alters weights according to loras. - """ - - lora_layer_name = getattr(self, 'lora_layer_name', None) - if lora_layer_name is None: - return - - current_names = getattr(self, "lora_current_names", ()) - wanted_names = tuple((x.name, x.multiplier) for x in loaded_loras) - - weights_backup = getattr(self, "lora_weights_backup", None) - if weights_backup is None: - if isinstance(self, torch.nn.MultiheadAttention): - weights_backup = (self.in_proj_weight.to(devices.cpu, copy=True), self.out_proj.weight.to(devices.cpu, copy=True)) - else: - weights_backup = self.weight.to(devices.cpu, copy=True) - - self.lora_weights_backup = weights_backup - - if current_names != wanted_names: - lora_restore_weights_from_backup(self) - - for lora in loaded_loras: - module = lora.modules.get(lora_layer_name, None) - if module is not None and hasattr(self, 'weight'): - self.weight += lora_calc_updown(lora, module, self.weight) - continue - - module_q = lora.modules.get(lora_layer_name + "_q_proj", None) - module_k = lora.modules.get(lora_layer_name + "_k_proj", None) - module_v = lora.modules.get(lora_layer_name + "_v_proj", None) - module_out = lora.modules.get(lora_layer_name + "_out_proj", None) - - if isinstance(self, torch.nn.MultiheadAttention) and module_q and module_k and module_v and module_out: - updown_q = lora_calc_updown(lora, module_q, self.in_proj_weight) - updown_k = lora_calc_updown(lora, module_k, self.in_proj_weight) - updown_v = lora_calc_updown(lora, module_v, self.in_proj_weight) - updown_qkv = torch.vstack([updown_q, updown_k, updown_v]) - - self.in_proj_weight += updown_qkv - self.out_proj.weight += lora_calc_updown(lora, module_out, self.out_proj.weight) - continue - - if module is None: - continue - - print(f'failed to calculate lora weights for layer {lora_layer_name}') - - self.lora_current_names = wanted_names - - -def lora_forward(module, input, original_forward): - """ - Old way of applying Lora by executing operations during layer's forward. - Stacking many loras this way results in big performance degradation. - """ - - if len(loaded_loras) == 0: - return original_forward(module, input) - - input = devices.cond_cast_unet(input) - - lora_restore_weights_from_backup(module) - lora_reset_cached_weight(module) - - res = original_forward(module, input) - - lora_layer_name = getattr(module, 'lora_layer_name', None) - for lora in loaded_loras: - module = lora.modules.get(lora_layer_name, None) - if module is None: - continue - - module.up.to(device=devices.device) - module.down.to(device=devices.device) - - res = res + module.up(module.down(input)) * lora.multiplier * (module.alpha / module.up.weight.shape[1] if module.alpha else 1.0) - - return res - - -def lora_reset_cached_weight(self: Union[torch.nn.Conv2d, torch.nn.Linear]): - self.lora_current_names = () - self.lora_weights_backup = None - - -def lora_Linear_forward(self, input): - if shared.opts.lora_functional: - return lora_forward(self, input, torch.nn.Linear_forward_before_lora) - - lora_apply_weights(self) - - return torch.nn.Linear_forward_before_lora(self, input) - - -def lora_Linear_load_state_dict(self, *args, **kwargs): - lora_reset_cached_weight(self) - - return torch.nn.Linear_load_state_dict_before_lora(self, *args, **kwargs) - - -def lora_Conv2d_forward(self, input): - if shared.opts.lora_functional: - return lora_forward(self, input, torch.nn.Conv2d_forward_before_lora) - - lora_apply_weights(self) - - return torch.nn.Conv2d_forward_before_lora(self, input) - - -def lora_Conv2d_load_state_dict(self, *args, **kwargs): - lora_reset_cached_weight(self) - - return torch.nn.Conv2d_load_state_dict_before_lora(self, *args, **kwargs) - - -def lora_MultiheadAttention_forward(self, *args, **kwargs): - lora_apply_weights(self) - - return torch.nn.MultiheadAttention_forward_before_lora(self, *args, **kwargs) - - -def lora_MultiheadAttention_load_state_dict(self, *args, **kwargs): - lora_reset_cached_weight(self) - - return torch.nn.MultiheadAttention_load_state_dict_before_lora(self, *args, **kwargs) - - -def list_available_loras(): - available_loras.clear() - available_lora_aliases.clear() - forbidden_lora_aliases.clear() - available_lora_hash_lookup.clear() - forbidden_lora_aliases.update({"none": 1, "Addams": 1}) - - os.makedirs(shared.cmd_opts.lora_dir, exist_ok=True) - - candidates = list(shared.walk_files(shared.cmd_opts.lora_dir, allowed_extensions=[".pt", ".ckpt", ".safetensors"])) - for filename in candidates: - if os.path.isdir(filename): - continue - - name = os.path.splitext(os.path.basename(filename))[0] - try: - entry = LoraOnDisk(name, filename) - except OSError: # should catch FileNotFoundError and PermissionError etc. - errors.report(f"Failed to load LoRA {name} from {filename}", exc_info=True) - continue - - available_loras[name] = entry - - if entry.alias in available_lora_aliases: - forbidden_lora_aliases[entry.alias.lower()] = 1 - - available_lora_aliases[name] = entry - available_lora_aliases[entry.alias] = entry - - -re_lora_name = re.compile(r"(.*)\s*\([0-9a-fA-F]+\)") - - -def infotext_pasted(infotext, params): - if "AddNet Module 1" in [x[1] for x in scripts.scripts_txt2img.infotext_fields]: - return # if the other extension is active, it will handle those fields, no need to do anything - - added = [] - - for k in params: - if not k.startswith("AddNet Model "): - continue - - num = k[13:] - - if params.get("AddNet Module " + num) != "LoRA": - continue - - name = params.get("AddNet Model " + num) - if name is None: - continue - - m = re_lora_name.match(name) - if m: - name = m.group(1) - - multiplier = params.get("AddNet Weight A " + num, "1.0") - - added.append(f"") - - if added: - params["Prompt"] += "\n" + "".join(added) - - -available_loras = {} -available_lora_aliases = {} -available_lora_hash_lookup = {} -forbidden_lora_aliases = {} -loaded_loras = [] - -list_available_loras() diff --git a/extensions-builtin/Lora/lyco_helpers.py b/extensions-builtin/Lora/lyco_helpers.py new file mode 100644 index 00000000..9ea499fb --- /dev/null +++ b/extensions-builtin/Lora/lyco_helpers.py @@ -0,0 +1,15 @@ +import torch + + +def make_weight_cp(t, wa, wb): + temp = torch.einsum('i j k l, j r -> i r k l', t, wb) + return torch.einsum('i j k l, i r -> r j k l', temp, wa) + + +def rebuild_conventional(up, down, shape, dyn_dim=None): + up = up.reshape(up.size(0), -1) + down = down.reshape(down.size(0), -1) + if dyn_dim is not None: + up = up[:, :dyn_dim] + down = down[:dyn_dim, :] + return (up @ down).reshape(shape) diff --git a/extensions-builtin/Lora/network.py b/extensions-builtin/Lora/network.py new file mode 100644 index 00000000..a1fe6bbf --- /dev/null +++ b/extensions-builtin/Lora/network.py @@ -0,0 +1,98 @@ +import os +from collections import namedtuple + +import torch + +from modules import devices, sd_models, cache, errors, hashes, shared + +NetworkWeights = namedtuple('NetworkWeights', ['network_key', 'sd_key', 'w', 'sd_module']) + +metadata_tags_order = {"ss_sd_model_name": 1, "ss_resolution": 2, "ss_clip_skip": 3, "ss_num_train_images": 10, "ss_tag_frequency": 20} + + +class NetworkOnDisk: + def __init__(self, name, filename): + self.name = name + self.filename = filename + self.metadata = {} + self.is_safetensors = os.path.splitext(filename)[1].lower() == ".safetensors" + + def read_metadata(): + metadata = sd_models.read_metadata_from_safetensors(filename) + metadata.pop('ssmd_cover_images', None) # those are cover images, and they are too big to display in UI as text + + return metadata + + if self.is_safetensors: + try: + self.metadata = cache.cached_data_for_file('safetensors-metadata', "lora/" + self.name, filename, read_metadata) + except Exception as e: + errors.display(e, f"reading lora {filename}") + + if self.metadata: + m = {} + for k, v in sorted(self.metadata.items(), key=lambda x: metadata_tags_order.get(x[0], 999)): + m[k] = v + + self.metadata = m + + self.alias = self.metadata.get('ss_output_name', self.name) + + self.hash = None + self.shorthash = None + self.set_hash( + self.metadata.get('sshs_model_hash') or + hashes.sha256_from_cache(self.filename, "lora/" + self.name, use_addnet_hash=self.is_safetensors) or + '' + ) + + def set_hash(self, v): + self.hash = v + self.shorthash = self.hash[0:12] + + if self.shorthash: + import networks + networks.available_network_hash_lookup[self.shorthash] = self + + def read_hash(self): + if not self.hash: + self.set_hash(hashes.sha256(self.filename, "lora/" + self.name, use_addnet_hash=self.is_safetensors) or '') + + def get_alias(self): + import networks + if shared.opts.lora_preferred_name == "Filename" or self.alias.lower() in networks.forbidden_network_aliases: + return self.name + else: + return self.alias + + +class Network: # LoraModule + def __init__(self, name, network_on_disk: NetworkOnDisk): + self.name = name + self.network_on_disk = network_on_disk + self.multiplier = 1.0 + self.modules = {} + self.mtime = None + + self.mentioned_name = None + """the text that was used to add the network to prompt - can be either name or an alias""" + + +class ModuleType: + def create_module(self, net: Network, weights: NetworkWeights) -> Network | None: + return None + + +class NetworkModule: + def __init__(self, net: Network, weights: NetworkWeights): + self.network = net + self.network_key = weights.network_key + self.sd_key = weights.sd_key + self.sd_module = weights.sd_module + + def calc_updown(self, target): + raise NotImplementedError() + + def forward(self, x, y): + raise NotImplementedError() + diff --git a/extensions-builtin/Lora/network_hada.py b/extensions-builtin/Lora/network_hada.py new file mode 100644 index 00000000..15e7ffd8 --- /dev/null +++ b/extensions-builtin/Lora/network_hada.py @@ -0,0 +1,59 @@ +import lyco_helpers +import network +import network_lyco + + +class ModuleTypeHada(network.ModuleType): + def create_module(self, net: network.Network, weights: network.NetworkWeights): + if all(x in weights.w for x in ["hada_w1_a", "hada_w1_b", "hada_w2_a", "hada_w2_b"]): + return NetworkModuleHada(net, weights) + + return None + + +class NetworkModuleHada(network_lyco.NetworkModuleLyco): + def __init__(self, net: network.Network, weights: network.NetworkWeights): + super().__init__(net, weights) + + if hasattr(self.sd_module, 'weight'): + self.shape = self.sd_module.weight.shape + + self.w1a = weights.w["hada_w1_a"] + self.w1b = weights.w["hada_w1_b"] + self.dim = self.w1b.shape[0] + self.w2a = weights.w["hada_w2_a"] + self.w2b = weights.w["hada_w2_b"] + + self.t1 = weights.w.get("hada_t1") + self.t2 = weights.w.get("hada_t2") + + self.alpha = weights.w["alpha"].item() if "alpha" in weights.w else None + self.scale = weights.w["scale"].item() if "scale" in weights.w else None + + def calc_updown(self, orig_weight): + w1a = self.w1a.to(orig_weight.device, dtype=orig_weight.dtype) + w1b = self.w1b.to(orig_weight.device, dtype=orig_weight.dtype) + w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype) + w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype) + + output_shape = [w1a.size(0), w1b.size(1)] + + if self.t1 is not None: + output_shape = [w1a.size(1), w1b.size(1)] + t1 = self.t1.to(orig_weight.device, dtype=orig_weight.dtype) + updown1 = lyco_helpers.make_weight_cp(t1, w1a, w1b) + output_shape += t1.shape[2:] + else: + if len(w1b.shape) == 4: + output_shape += w1b.shape[2:] + updown1 = lyco_helpers.rebuild_conventional(w1a, w1b, output_shape) + + if self.t2 is not None: + t2 = self.t2.to(orig_weight.device, dtype=orig_weight.dtype) + updown2 = lyco_helpers.make_weight_cp(t2, w2a, w2b) + else: + updown2 = lyco_helpers.rebuild_conventional(w2a, w2b, output_shape) + + updown = updown1 * updown2 + + return self.finalize_updown(updown, orig_weight, output_shape) diff --git a/extensions-builtin/Lora/network_lora.py b/extensions-builtin/Lora/network_lora.py new file mode 100644 index 00000000..b2d96537 --- /dev/null +++ b/extensions-builtin/Lora/network_lora.py @@ -0,0 +1,70 @@ +import torch + +import network +from modules import devices + + +class ModuleTypeLora(network.ModuleType): + def create_module(self, net: network.Network, weights: network.NetworkWeights): + if all(x in weights.w for x in ["lora_up.weight", "lora_down.weight"]): + return NetworkModuleLora(net, weights) + + return None + + +class NetworkModuleLora(network.NetworkModule): + def __init__(self, net: network.Network, weights: network.NetworkWeights): + super().__init__(net, weights) + + self.up = self.create_module(weights.w["lora_up.weight"]) + self.down = self.create_module(weights.w["lora_down.weight"]) + self.alpha = weights.w["alpha"] if "alpha" in weights.w else None + + def create_module(self, weight, none_ok=False): + if weight is None and none_ok: + return None + + if type(self.sd_module) == torch.nn.Linear: + module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) + elif type(self.sd_module) == torch.nn.modules.linear.NonDynamicallyQuantizableLinear: + module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) + elif type(self.sd_module) == torch.nn.MultiheadAttention: + module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) + elif type(self.sd_module) == torch.nn.Conv2d and weight.shape[2:] == (1, 1): + module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (1, 1), bias=False) + elif type(self.sd_module) == torch.nn.Conv2d and weight.shape[2:] == (3, 3): + module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (3, 3), bias=False) + else: + print(f'Network layer {self.network_key} matched a layer with unsupported type: {type(self.sd_module).__name__}') + return None + + with torch.no_grad(): + module.weight.copy_(weight) + + module.to(device=devices.cpu, dtype=devices.dtype) + module.weight.requires_grad_(False) + + return module + + def calc_updown(self, target): + up = self.up.weight.to(target.device, dtype=target.dtype) + down = self.down.weight.to(target.device, dtype=target.dtype) + + if up.shape[2:] == (1, 1) and down.shape[2:] == (1, 1): + updown = (up.squeeze(2).squeeze(2) @ down.squeeze(2).squeeze(2)).unsqueeze(2).unsqueeze(3) + elif up.shape[2:] == (3, 3) or down.shape[2:] == (3, 3): + updown = torch.nn.functional.conv2d(down.permute(1, 0, 2, 3), up).permute(1, 0, 2, 3) + else: + updown = up @ down + + updown = updown * self.network.multiplier * (self.alpha / self.up.weight.shape[1] if self.alpha else 1.0) + + return updown + + def forward(self, x, y): + self.up.to(device=devices.device) + self.down.to(device=devices.device) + + return y + self.up(self.down(x)) * self.network.multiplier * (self.alpha / self.up.weight.shape[1] if self.alpha else 1.0) + + diff --git a/extensions-builtin/Lora/network_lyco.py b/extensions-builtin/Lora/network_lyco.py new file mode 100644 index 00000000..18a822fa --- /dev/null +++ b/extensions-builtin/Lora/network_lyco.py @@ -0,0 +1,39 @@ +import torch + +import lyco_helpers +import network +from modules import devices + + +class NetworkModuleLyco(network.NetworkModule): + def __init__(self, net: network.Network, weights: network.NetworkWeights): + super().__init__(net, weights) + + if hasattr(self.sd_module, 'weight'): + self.shape = self.sd_module.weight.shape + + self.dim = None + self.bias = weights.w.get("bias") + self.alpha = weights.w["alpha"].item() if "alpha" in weights.w else None + self.scale = weights.w["scale"].item() if "scale" in weights.w else None + + def finalize_updown(self, updown, orig_weight, output_shape): + if self.bias is not None: + updown = updown.reshape(self.bias.shape) + updown += self.bias.to(orig_weight.device, dtype=orig_weight.dtype) + updown = updown.reshape(output_shape) + + if len(output_shape) == 4: + updown = updown.reshape(output_shape) + + if orig_weight.size().numel() == updown.size().numel(): + updown = updown.reshape(orig_weight.shape) + + scale = ( + self.scale if self.scale is not None + else self.alpha / self.dim if self.dim is not None and self.alpha is not None + else 1.0 + ) + + return updown * scale * self.network.multiplier + diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py new file mode 100644 index 00000000..5b0ddfb6 --- /dev/null +++ b/extensions-builtin/Lora/networks.py @@ -0,0 +1,443 @@ +import os +import re + +import network +import network_lora +import network_hada + +import torch +from typing import Union + +from modules import shared, devices, sd_models, errors, scripts, sd_hijack + +module_types = [ + network_lora.ModuleTypeLora(), + network_hada.ModuleTypeHada(), +] + + +re_digits = re.compile(r"\d+") +re_x_proj = re.compile(r"(.*)_([qkv]_proj)$") +re_compiled = {} + +suffix_conversion = { + "attentions": {}, + "resnets": { + "conv1": "in_layers_2", + "conv2": "out_layers_3", + "time_emb_proj": "emb_layers_1", + "conv_shortcut": "skip_connection", + } +} + + +def convert_diffusers_name_to_compvis(key, is_sd2): + def match(match_list, regex_text): + regex = re_compiled.get(regex_text) + if regex is None: + regex = re.compile(regex_text) + re_compiled[regex_text] = regex + + r = re.match(regex, key) + if not r: + return False + + match_list.clear() + match_list.extend([int(x) if re.match(re_digits, x) else x for x in r.groups()]) + return True + + m = [] + + if match(m, r"lora_unet_down_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"): + suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3]) + return f"diffusion_model_input_blocks_{1 + m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}" + + if match(m, r"lora_unet_mid_block_(attentions|resnets)_(\d+)_(.+)"): + suffix = suffix_conversion.get(m[0], {}).get(m[2], m[2]) + return f"diffusion_model_middle_block_{1 if m[0] == 'attentions' else m[1] * 2}_{suffix}" + + if match(m, r"lora_unet_up_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"): + suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3]) + return f"diffusion_model_output_blocks_{m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}" + + if match(m, r"lora_unet_down_blocks_(\d+)_downsamplers_0_conv"): + return f"diffusion_model_input_blocks_{3 + m[0] * 3}_0_op" + + if match(m, r"lora_unet_up_blocks_(\d+)_upsamplers_0_conv"): + return f"diffusion_model_output_blocks_{2 + m[0] * 3}_{2 if m[0]>0 else 1}_conv" + + if match(m, r"lora_te_text_model_encoder_layers_(\d+)_(.+)"): + if is_sd2: + if 'mlp_fc1' in m[1]: + return f"model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc1', 'mlp_c_fc')}" + elif 'mlp_fc2' in m[1]: + return f"model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc2', 'mlp_c_proj')}" + else: + return f"model_transformer_resblocks_{m[0]}_{m[1].replace('self_attn', 'attn')}" + + return f"transformer_text_model_encoder_layers_{m[0]}_{m[1]}" + + if match(m, r"lora_te2_text_model_encoder_layers_(\d+)_(.+)"): + if 'mlp_fc1' in m[1]: + return f"1_model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc1', 'mlp_c_fc')}" + elif 'mlp_fc2' in m[1]: + return f"1_model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc2', 'mlp_c_proj')}" + else: + return f"1_model_transformer_resblocks_{m[0]}_{m[1].replace('self_attn', 'attn')}" + + return key + + +def assign_network_names_to_compvis_modules(sd_model): + network_layer_mapping = {} + + if shared.sd_model.is_sdxl: + for i, embedder in enumerate(shared.sd_model.conditioner.embedders): + if not hasattr(embedder, 'wrapped'): + continue + + for name, module in embedder.wrapped.named_modules(): + network_name = f'{i}_{name.replace(".", "_")}' + network_layer_mapping[network_name] = module + module.network_layer_name = network_name + else: + for name, module in shared.sd_model.cond_stage_model.wrapped.named_modules(): + network_name = name.replace(".", "_") + network_layer_mapping[network_name] = module + module.network_layer_name = network_name + + for name, module in shared.sd_model.model.named_modules(): + network_name = name.replace(".", "_") + network_layer_mapping[network_name] = module + module.network_layer_name = network_name + + sd_model.network_layer_mapping = network_layer_mapping + + +def load_network(name, network_on_disk): + net = network.Network(name, network_on_disk) + net.mtime = os.path.getmtime(network_on_disk.filename) + + sd = sd_models.read_state_dict(network_on_disk.filename) + + # this should not be needed but is here as an emergency fix for an unknown error people are experiencing in 1.2.0 + if not hasattr(shared.sd_model, 'network_layer_mapping'): + assign_network_names_to_compvis_modules(shared.sd_model) + + keys_failed_to_match = {} + is_sd2 = 'model_transformer_resblocks' in shared.sd_model.network_layer_mapping + + matched_networks = {} + + for key_network, weight in sd.items(): + key_network_without_network_parts, network_part = key_network.split(".", 1) + + key = convert_diffusers_name_to_compvis(key_network_without_network_parts, is_sd2) + sd_module = shared.sd_model.network_layer_mapping.get(key, None) + + if sd_module is None: + m = re_x_proj.match(key) + if m: + sd_module = shared.sd_model.network_layer_mapping.get(m.group(1), None) + + # SDXL loras seem to already have correct compvis keys, so only need to replace "lora_unet" with "diffusion_model" + if sd_module is None and "lora_unet" in key_network_without_network_parts: + key = key_network_without_network_parts.replace("lora_unet", "diffusion_model") + sd_module = shared.sd_model.network_layer_mapping.get(key, None) + elif sd_module is None and "lora_te1_text_model" in key_network_without_network_parts: + key = key_network_without_network_parts.replace("lora_te1_text_model", "0_transformer_text_model") + sd_module = shared.sd_model.network_layer_mapping.get(key, None) + + if sd_module is None: + keys_failed_to_match[key_network] = key + continue + + if key not in matched_networks: + matched_networks[key] = network.NetworkWeights(network_key=key_network, sd_key=key, w={}, sd_module=sd_module) + + matched_networks[key].w[network_part] = weight + + for key, weights in matched_networks.items(): + net_module = None + for nettype in module_types: + net_module = nettype.create_module(net, weights) + if net_module is not None: + break + + if net_module is None: + raise AssertionError(f"Could not find a module type (out of {', '.join([x.__class__.__name__ for x in module_types])}) that would accept those keys: {', '.join(weights.w)}") + + 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}") + + return net + + +def load_networks(names, multipliers=None): + already_loaded = {} + + for net in loaded_networks: + if net.name in names: + already_loaded[net.name] = net + + loaded_networks.clear() + + networks_on_disk = [available_network_aliases.get(name, None) for name in names] + if any(x is None for x in networks_on_disk): + list_available_networks() + + networks_on_disk = [available_network_aliases.get(name, None) for name in names] + + failed_to_load_networks = [] + + for i, name in enumerate(names): + net = already_loaded.get(name, None) + + network_on_disk = networks_on_disk[i] + + if network_on_disk is not None: + if net is None or os.path.getmtime(network_on_disk.filename) > net.mtime: + try: + net = load_network(name, network_on_disk) + except Exception as e: + errors.display(e, f"loading network {network_on_disk.filename}") + continue + + net.mentioned_name = name + + network_on_disk.read_hash() + + if net is None: + failed_to_load_networks.append(name) + print(f"Couldn't find network with name {name}") + continue + + net.multiplier = multipliers[i] if multipliers else 1.0 + loaded_networks.append(net) + + if failed_to_load_networks: + sd_hijack.model_hijack.comments.append("Failed to find networks: " + ", ".join(failed_to_load_networks)) + + +def network_restore_weights_from_backup(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn.MultiheadAttention]): + weights_backup = getattr(self, "network_weights_backup", None) + + if weights_backup is None: + return + + if isinstance(self, torch.nn.MultiheadAttention): + self.in_proj_weight.copy_(weights_backup[0]) + self.out_proj.weight.copy_(weights_backup[1]) + else: + self.weight.copy_(weights_backup) + + +def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn.MultiheadAttention]): + """ + Applies the currently selected set of networks to the weights of torch layer self. + If weights already have this particular set of networks applied, does nothing. + If not, restores orginal weights from backup and alters weights according to networks. + """ + + network_layer_name = getattr(self, 'network_layer_name', None) + if network_layer_name is None: + return + + current_names = getattr(self, "network_current_names", ()) + wanted_names = tuple((x.name, x.multiplier) for x in loaded_networks) + + weights_backup = getattr(self, "network_weights_backup", None) + if weights_backup is None: + if isinstance(self, torch.nn.MultiheadAttention): + weights_backup = (self.in_proj_weight.to(devices.cpu, copy=True), self.out_proj.weight.to(devices.cpu, copy=True)) + else: + weights_backup = self.weight.to(devices.cpu, copy=True) + + self.network_weights_backup = weights_backup + + if current_names != wanted_names: + network_restore_weights_from_backup(self) + + 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 = 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)) + + self.weight += updown + + module_q = net.modules.get(network_layer_name + "_q_proj", None) + module_k = net.modules.get(network_layer_name + "_k_proj", None) + module_v = net.modules.get(network_layer_name + "_v_proj", None) + 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]) + + self.in_proj_weight += updown_qkv + self.out_proj.weight += module_out.calc_updown(self.out_proj.weight) + continue + + if module is None: + continue + + print(f'failed to calculate network weights for layer {network_layer_name}') + + self.network_current_names = wanted_names + + +def network_forward(module, input, original_forward): + """ + Old way of applying Lora by executing operations during layer's forward. + Stacking many loras this way results in big performance degradation. + """ + + if len(loaded_networks) == 0: + return original_forward(module, input) + + input = devices.cond_cast_unet(input) + + network_restore_weights_from_backup(module) + network_reset_cached_weight(module) + + y = original_forward(module, input) + + network_layer_name = getattr(module, 'network_layer_name', None) + for lora in loaded_networks: + module = lora.modules.get(network_layer_name, None) + if module is None: + continue + + y = module.forward(y, input) + + return y + + +def network_reset_cached_weight(self: Union[torch.nn.Conv2d, torch.nn.Linear]): + self.network_current_names = () + self.network_weights_backup = None + + +def network_Linear_forward(self, input): + if shared.opts.lora_functional: + return network_forward(self, input, torch.nn.Linear_forward_before_network) + + network_apply_weights(self) + + return torch.nn.Linear_forward_before_network(self, input) + + +def network_Linear_load_state_dict(self, *args, **kwargs): + network_reset_cached_weight(self) + + return torch.nn.Linear_load_state_dict_before_network(self, *args, **kwargs) + + +def network_Conv2d_forward(self, input): + if shared.opts.lora_functional: + return network_forward(self, input, torch.nn.Conv2d_forward_before_network) + + network_apply_weights(self) + + return torch.nn.Conv2d_forward_before_network(self, input) + + +def network_Conv2d_load_state_dict(self, *args, **kwargs): + network_reset_cached_weight(self) + + return torch.nn.Conv2d_load_state_dict_before_network(self, *args, **kwargs) + + +def network_MultiheadAttention_forward(self, *args, **kwargs): + network_apply_weights(self) + + return torch.nn.MultiheadAttention_forward_before_network(self, *args, **kwargs) + + +def network_MultiheadAttention_load_state_dict(self, *args, **kwargs): + network_reset_cached_weight(self) + + return torch.nn.MultiheadAttention_load_state_dict_before_network(self, *args, **kwargs) + + +def list_available_networks(): + available_networks.clear() + available_network_aliases.clear() + forbidden_network_aliases.clear() + available_network_hash_lookup.clear() + forbidden_network_aliases.update({"none": 1, "Addams": 1}) + + os.makedirs(shared.cmd_opts.lora_dir, exist_ok=True) + + candidates = list(shared.walk_files(shared.cmd_opts.lora_dir, allowed_extensions=[".pt", ".ckpt", ".safetensors"])) + for filename in candidates: + if os.path.isdir(filename): + continue + + name = os.path.splitext(os.path.basename(filename))[0] + try: + entry = network.NetworkOnDisk(name, filename) + except OSError: # should catch FileNotFoundError and PermissionError etc. + errors.report(f"Failed to load network {name} from {filename}", exc_info=True) + continue + + available_networks[name] = entry + + if entry.alias in available_network_aliases: + forbidden_network_aliases[entry.alias.lower()] = 1 + + available_network_aliases[name] = entry + available_network_aliases[entry.alias] = entry + + +re_network_name = re.compile(r"(.*)\s*\([0-9a-fA-F]+\)") + + +def infotext_pasted(infotext, params): + if "AddNet Module 1" in [x[1] for x in scripts.scripts_txt2img.infotext_fields]: + return # if the other extension is active, it will handle those fields, no need to do anything + + added = [] + + for k in params: + if not k.startswith("AddNet Model "): + continue + + num = k[13:] + + if params.get("AddNet Module " + num) != "LoRA": + continue + + name = params.get("AddNet Model " + num) + if name is None: + continue + + m = re_network_name.match(name) + if m: + name = m.group(1) + + multiplier = params.get("AddNet Weight A " + num, "1.0") + + added.append(f"") + + if added: + params["Prompt"] += "\n" + "".join(added) + + +available_networks = {} +available_network_aliases = {} +loaded_networks = [] +available_network_hash_lookup = {} +forbidden_network_aliases = {} + +list_available_networks() diff --git a/extensions-builtin/Lora/scripts/lora_script.py b/extensions-builtin/Lora/scripts/lora_script.py index e650f469..81e6572a 100644 --- a/extensions-builtin/Lora/scripts/lora_script.py +++ b/extensions-builtin/Lora/scripts/lora_script.py @@ -4,18 +4,19 @@ import torch import gradio as gr from fastapi import FastAPI -import lora +import network +import networks import extra_networks_lora import ui_extra_networks_lora from modules import script_callbacks, ui_extra_networks, extra_networks, shared def unload(): - torch.nn.Linear.forward = torch.nn.Linear_forward_before_lora - torch.nn.Linear._load_from_state_dict = torch.nn.Linear_load_state_dict_before_lora - torch.nn.Conv2d.forward = torch.nn.Conv2d_forward_before_lora - torch.nn.Conv2d._load_from_state_dict = torch.nn.Conv2d_load_state_dict_before_lora - torch.nn.MultiheadAttention.forward = torch.nn.MultiheadAttention_forward_before_lora - torch.nn.MultiheadAttention._load_from_state_dict = torch.nn.MultiheadAttention_load_state_dict_before_lora + torch.nn.Linear.forward = torch.nn.Linear_forward_before_network + torch.nn.Linear._load_from_state_dict = torch.nn.Linear_load_state_dict_before_network + torch.nn.Conv2d.forward = torch.nn.Conv2d_forward_before_network + torch.nn.Conv2d._load_from_state_dict = torch.nn.Conv2d_load_state_dict_before_network + torch.nn.MultiheadAttention.forward = torch.nn.MultiheadAttention_forward_before_network + torch.nn.MultiheadAttention._load_from_state_dict = torch.nn.MultiheadAttention_load_state_dict_before_network def before_ui(): @@ -23,50 +24,50 @@ def before_ui(): extra_networks.register_extra_network(extra_networks_lora.ExtraNetworkLora()) -if not hasattr(torch.nn, 'Linear_forward_before_lora'): - torch.nn.Linear_forward_before_lora = torch.nn.Linear.forward +if not hasattr(torch.nn, 'Linear_forward_before_network'): + torch.nn.Linear_forward_before_network = torch.nn.Linear.forward -if not hasattr(torch.nn, 'Linear_load_state_dict_before_lora'): - torch.nn.Linear_load_state_dict_before_lora = torch.nn.Linear._load_from_state_dict +if not hasattr(torch.nn, 'Linear_load_state_dict_before_network'): + torch.nn.Linear_load_state_dict_before_network = torch.nn.Linear._load_from_state_dict -if not hasattr(torch.nn, 'Conv2d_forward_before_lora'): - torch.nn.Conv2d_forward_before_lora = torch.nn.Conv2d.forward +if not hasattr(torch.nn, 'Conv2d_forward_before_network'): + torch.nn.Conv2d_forward_before_network = torch.nn.Conv2d.forward -if not hasattr(torch.nn, 'Conv2d_load_state_dict_before_lora'): - torch.nn.Conv2d_load_state_dict_before_lora = torch.nn.Conv2d._load_from_state_dict +if not hasattr(torch.nn, 'Conv2d_load_state_dict_before_network'): + torch.nn.Conv2d_load_state_dict_before_network = torch.nn.Conv2d._load_from_state_dict -if not hasattr(torch.nn, 'MultiheadAttention_forward_before_lora'): - torch.nn.MultiheadAttention_forward_before_lora = torch.nn.MultiheadAttention.forward +if not hasattr(torch.nn, 'MultiheadAttention_forward_before_network'): + torch.nn.MultiheadAttention_forward_before_network = torch.nn.MultiheadAttention.forward -if not hasattr(torch.nn, 'MultiheadAttention_load_state_dict_before_lora'): - torch.nn.MultiheadAttention_load_state_dict_before_lora = torch.nn.MultiheadAttention._load_from_state_dict +if not hasattr(torch.nn, 'MultiheadAttention_load_state_dict_before_network'): + torch.nn.MultiheadAttention_load_state_dict_before_network = torch.nn.MultiheadAttention._load_from_state_dict -torch.nn.Linear.forward = lora.lora_Linear_forward -torch.nn.Linear._load_from_state_dict = lora.lora_Linear_load_state_dict -torch.nn.Conv2d.forward = lora.lora_Conv2d_forward -torch.nn.Conv2d._load_from_state_dict = lora.lora_Conv2d_load_state_dict -torch.nn.MultiheadAttention.forward = lora.lora_MultiheadAttention_forward -torch.nn.MultiheadAttention._load_from_state_dict = lora.lora_MultiheadAttention_load_state_dict +torch.nn.Linear.forward = networks.network_Linear_forward +torch.nn.Linear._load_from_state_dict = networks.network_Linear_load_state_dict +torch.nn.Conv2d.forward = networks.network_Conv2d_forward +torch.nn.Conv2d._load_from_state_dict = networks.network_Conv2d_load_state_dict +torch.nn.MultiheadAttention.forward = networks.network_MultiheadAttention_forward +torch.nn.MultiheadAttention._load_from_state_dict = networks.network_MultiheadAttention_load_state_dict -script_callbacks.on_model_loaded(lora.assign_lora_names_to_compvis_modules) +script_callbacks.on_model_loaded(networks.assign_network_names_to_compvis_modules) script_callbacks.on_script_unloaded(unload) script_callbacks.on_before_ui(before_ui) -script_callbacks.on_infotext_pasted(lora.infotext_pasted) +script_callbacks.on_infotext_pasted(networks.infotext_pasted) shared.options_templates.update(shared.options_section(('extra_networks', "Extra Networks"), { - "sd_lora": shared.OptionInfo("None", "Add Lora to prompt", gr.Dropdown, lambda: {"choices": ["None", *lora.available_loras]}, refresh=lora.list_available_loras), + "sd_lora": shared.OptionInfo("None", "Add network to prompt", gr.Dropdown, lambda: {"choices": ["None", *networks.available_networks]}, refresh=networks.list_available_networks), "lora_preferred_name": shared.OptionInfo("Alias from file", "When adding to prompt, refer to Lora by", gr.Radio, {"choices": ["Alias from file", "Filename"]}), "lora_add_hashes_to_infotext": shared.OptionInfo(True, "Add Lora hashes to infotext"), })) shared.options_templates.update(shared.options_section(('compatibility', "Compatibility"), { - "lora_functional": shared.OptionInfo(False, "Lora: use old method that takes longer when you have multiple Loras active and produces same results as kohya-ss/sd-webui-additional-networks extension"), + "lora_functional": shared.OptionInfo(False, "Lora/Networks: use old method that takes longer when you have multiple Loras active and produces same results as kohya-ss/sd-webui-additional-networks extension"), })) -def create_lora_json(obj: lora.LoraOnDisk): +def create_lora_json(obj: network.NetworkOnDisk): return { "name": obj.name, "alias": obj.alias, @@ -75,17 +76,17 @@ def create_lora_json(obj: lora.LoraOnDisk): } -def api_loras(_: gr.Blocks, app: FastAPI): +def api_networks(_: gr.Blocks, app: FastAPI): @app.get("/sdapi/v1/loras") async def get_loras(): - return [create_lora_json(obj) for obj in lora.available_loras.values()] + return [create_lora_json(obj) for obj in networks.available_networks.values()] @app.post("/sdapi/v1/refresh-loras") async def refresh_loras(): - return lora.list_available_loras() + return networks.list_available_networks() -script_callbacks.on_app_started(api_loras) +script_callbacks.on_app_started(api_networks) re_lora = re.compile(" Date: Sun, 16 Jul 2023 23:14:57 +0300 Subject: [PATCH 003/147] linter --- extensions-builtin/Lora/network.py | 4 +--- extensions-builtin/Lora/network_lyco.py | 4 ---- 2 files changed, 1 insertion(+), 7 deletions(-) diff --git a/extensions-builtin/Lora/network.py b/extensions-builtin/Lora/network.py index a1fe6bbf..4ac63722 100644 --- a/extensions-builtin/Lora/network.py +++ b/extensions-builtin/Lora/network.py @@ -1,9 +1,7 @@ import os from collections import namedtuple -import torch - -from modules import devices, sd_models, cache, errors, hashes, shared +from modules import sd_models, cache, errors, hashes, shared NetworkWeights = namedtuple('NetworkWeights', ['network_key', 'sd_key', 'w', 'sd_module']) diff --git a/extensions-builtin/Lora/network_lyco.py b/extensions-builtin/Lora/network_lyco.py index 18a822fa..fc135314 100644 --- a/extensions-builtin/Lora/network_lyco.py +++ b/extensions-builtin/Lora/network_lyco.py @@ -1,8 +1,4 @@ -import torch - -import lyco_helpers import network -from modules import devices class NetworkModuleLyco(network.NetworkModule): From ef5dac7786916dd39711edb2b8e90ce96ef78fca Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 00:01:17 +0300 Subject: [PATCH 004/147] fix --- extensions-builtin/Lora/network_hada.py | 3 --- extensions-builtin/Lora/networks.py | 1 + 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/extensions-builtin/Lora/network_hada.py b/extensions-builtin/Lora/network_hada.py index 15e7ffd8..799bb3bc 100644 --- a/extensions-builtin/Lora/network_hada.py +++ b/extensions-builtin/Lora/network_hada.py @@ -27,9 +27,6 @@ class NetworkModuleHada(network_lyco.NetworkModuleLyco): self.t1 = weights.w.get("hada_t1") self.t2 = weights.w.get("hada_t2") - self.alpha = weights.w["alpha"].item() if "alpha" in weights.w else None - self.scale = weights.w["scale"].item() if "scale" in weights.w else None - def calc_updown(self, orig_weight): w1a = self.w1a.to(orig_weight.device, dtype=orig_weight.dtype) w1b = self.w1b.to(orig_weight.device, dtype=orig_weight.dtype) diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index 5b0ddfb6..90374faa 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -271,6 +271,7 @@ def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn updown = torch.nn.functional.pad(updown, (0, 0, 0, 0, 0, 5)) self.weight += updown + continue module_q = net.modules.get(network_layer_name + "_q_proj", None) module_k = net.modules.get(network_layer_name + "_k_proj", None) From 58c3df32f3a73b20ea33d1709a1d25818b8a98dd Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 00:12:18 +0300 Subject: [PATCH 005/147] IA3 support --- extensions-builtin/Lora/network_ia3.py | 32 ++++++++++++++++++++++++++ extensions-builtin/Lora/networks.py | 2 ++ 2 files changed, 34 insertions(+) create mode 100644 extensions-builtin/Lora/network_ia3.py diff --git a/extensions-builtin/Lora/network_ia3.py b/extensions-builtin/Lora/network_ia3.py new file mode 100644 index 00000000..99f2307c --- /dev/null +++ b/extensions-builtin/Lora/network_ia3.py @@ -0,0 +1,32 @@ +import lyco_helpers +import network +import network_lyco + + +class ModuleTypeIa3(network.ModuleType): + def create_module(self, net: network.Network, weights: network.NetworkWeights): + if all(x in weights.w for x in ["weight"]): + return NetworkModuleIa3(net, weights) + + return None + + +class NetworkModuleIa3(network_lyco.NetworkModuleLyco): + def __init__(self, net: network.Network, weights: network.NetworkWeights): + super().__init__(net, weights) + + self.w = weights.w["weight"] + self.on_input = weights.w["on_input"].item() + + def calc_updown(self, orig_weight): + w = self.w.to(orig_weight.device, dtype=orig_weight.dtype) + + output_shape = [w.size(0), orig_weight.size(1)] + if self.on_input: + output_shape.reverse() + else: + w = w.reshape(-1, 1) + + updown = orig_weight * w + + return self.finalize_updown(updown, orig_weight, output_shape) diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index 90374faa..bf810b5b 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -4,6 +4,7 @@ import re import network import network_lora import network_hada +import network_ia3 import torch from typing import Union @@ -13,6 +14,7 @@ from modules import shared, devices, sd_models, errors, scripts, sd_hijack module_types = [ network_lora.ModuleTypeLora(), network_hada.ModuleTypeHada(), + network_ia3.ModuleTypeIa3(), ] From 46466f09d0b0c14118033dee6af0f876059776d3 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 00:29:07 +0300 Subject: [PATCH 006/147] Lokr support --- extensions-builtin/Lora/network_ia3.py | 1 - extensions-builtin/Lora/network_lokr.py | 65 +++++++++++++++++++++++++ extensions-builtin/Lora/networks.py | 2 + 3 files changed, 67 insertions(+), 1 deletion(-) create mode 100644 extensions-builtin/Lora/network_lokr.py diff --git a/extensions-builtin/Lora/network_ia3.py b/extensions-builtin/Lora/network_ia3.py index 99f2307c..d8806da0 100644 --- a/extensions-builtin/Lora/network_ia3.py +++ b/extensions-builtin/Lora/network_ia3.py @@ -1,4 +1,3 @@ -import lyco_helpers import network import network_lyco diff --git a/extensions-builtin/Lora/network_lokr.py b/extensions-builtin/Lora/network_lokr.py new file mode 100644 index 00000000..f1731924 --- /dev/null +++ b/extensions-builtin/Lora/network_lokr.py @@ -0,0 +1,65 @@ +import torch + +import lyco_helpers +import network +import network_lyco + + +class ModuleTypeLokr(network.ModuleType): + def create_module(self, net: network.Network, weights: network.NetworkWeights): + has_1 = "lokr_w1" in weights.w or ("lokr_w1a" in weights.w and "lokr_w1b" in weights.w) + has_2 = "lokr_w2" in weights.w or ("lokr_w2a" in weights.w and "lokr_w2b" in weights.w) + if has_1 and has_2: + return NetworkModuleLokr(net, weights) + + return None + + +def make_kron(orig_shape, w1, w2): + if len(w2.shape) == 4: + w1 = w1.unsqueeze(2).unsqueeze(2) + w2 = w2.contiguous() + return torch.kron(w1, w2).reshape(orig_shape) + + +class NetworkModuleLokr(network_lyco.NetworkModuleLyco): + def __init__(self, net: network.Network, weights: network.NetworkWeights): + super().__init__(net, weights) + + self.w1 = weights.w.get("lokr_w1") + self.w1a = weights.w.get("lokr_w1_a") + self.w1b = weights.w.get("lokr_w1_b") + self.dim = self.w1b.shape[0] if self.w1b else self.dim + self.w2 = weights.w.get("lokr_w2") + self.w2a = weights.w.get("lokr_w2_a") + self.w2b = weights.w.get("lokr_w2_b") + self.dim = self.w2b.shape[0] if self.w2b else self.dim + self.t2 = weights.w.get("lokr_t2") + + def calc_updown(self, orig_weight): + if self.w1 is not None: + w1 = self.w1.to(orig_weight.device, dtype=orig_weight.dtype) + else: + w1a = self.w1a.to(orig_weight.device, dtype=orig_weight.dtype) + w1b = self.w1b.to(orig_weight.device, dtype=orig_weight.dtype) + w1 = w1a @ w1b + + if self.w2 is not None: + w2 = self.w2.to(orig_weight.device, dtype=orig_weight.dtype) + elif self.t2 is None: + w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype) + w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype) + w2 = w2a @ w2b + else: + t2 = self.t2.to(orig_weight.device, dtype=orig_weight.dtype) + w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype) + w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype) + w2 = lyco_helpers.make_weight_cp(t2, w2a, w2b) + + output_shape = [w1.size(0) * w2.size(0), w1.size(1) * w2.size(1)] + if len(orig_weight.shape) == 4: + output_shape = orig_weight.shape + + updown = make_kron(output_shape, w1, w2) + + return self.finalize_updown(updown, orig_weight, output_shape) diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index bf810b5b..1b358561 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -5,6 +5,7 @@ import network import network_lora import network_hada import network_ia3 +import network_lokr import torch from typing import Union @@ -15,6 +16,7 @@ module_types = [ network_lora.ModuleTypeLora(), network_hada.ModuleTypeHada(), network_ia3.ModuleTypeIa3(), + network_lokr.ModuleTypeLokr(), ] From 7870937c770aaba9e681c299f923ba645163c85c Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Mon, 17 Jul 2023 12:25:29 +0900 Subject: [PATCH 007/147] XYZ always_discard_next_to_last_sigma Co-authored-by: Franck Mahon --- scripts/xyz_grid.py | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/scripts/xyz_grid.py b/scripts/xyz_grid.py index 7821cc65..ee30747c 100644 --- a/scripts/xyz_grid.py +++ b/scripts/xyz_grid.py @@ -144,11 +144,18 @@ def apply_face_restore(p, opt, x): p.restore_faces = is_active -def apply_override(field): +def apply_override(field, boolean: bool = False): def fun(p, x, xs): + if boolean: + x = True if x == "True" else False p.override_settings[field] = x return fun + +def boolean_choice(): + return ["True", "False"] + + def format_value_add_label(p, opt, x): if type(x) == float: x = round(x, 8) @@ -235,6 +242,7 @@ axis_options = [ AxisOption("Face restore", str, apply_face_restore, format_value=format_value), 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), ] From c03856bfdf30fd0e061caefd60231eb86a983c71 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Mon, 17 Jul 2023 12:45:10 +0900 Subject: [PATCH 008/147] reversible boolean_choice order --- scripts/xyz_grid.py | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/scripts/xyz_grid.py b/scripts/xyz_grid.py index ee30747c..bddc28c7 100644 --- a/scripts/xyz_grid.py +++ b/scripts/xyz_grid.py @@ -152,8 +152,10 @@ def apply_override(field, boolean: bool = False): return fun -def boolean_choice(): - return ["True", "False"] +def boolean_choice(reverse: bool = False): + def choice(): + return ["False", "True"] if reverse else ["True", "False"] + return choice def format_value_add_label(p, opt, x): @@ -242,7 +244,7 @@ axis_options = [ AxisOption("Face restore", str, apply_face_restore, format_value=format_value), 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), + AxisOption("Always discard next-to-last sigma", str, apply_override('always_discard_next_to_last_sigma', boolean=True), choices=boolean_choice(reverse=True)), ] From 8941297ceb3e71fa16fd842b135786b0ebc1b2b1 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Mon, 17 Jul 2023 12:45:38 +0900 Subject: [PATCH 009/147] lowercase --- scripts/xyz_grid.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/xyz_grid.py b/scripts/xyz_grid.py index bddc28c7..1010845e 100644 --- a/scripts/xyz_grid.py +++ b/scripts/xyz_grid.py @@ -147,7 +147,7 @@ def apply_face_restore(p, opt, x): def apply_override(field, boolean: bool = False): def fun(p, x, xs): if boolean: - x = True if x == "True" else False + x = True if x.lower() == "true" else False p.override_settings[field] = x return fun From 238adeaffb037dedbcefe41e7fd4814a1f17baa2 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 09:00:47 +0300 Subject: [PATCH 010/147] support specifying te and unet weights separately update lora code support full module --- .../Lora/extra_networks_lora.py | 22 ++++-- extensions-builtin/Lora/lyco_helpers.py | 6 ++ extensions-builtin/Lora/network.py | 40 ++++++++++- extensions-builtin/Lora/network_full.py | 23 ++++++ extensions-builtin/Lora/network_hada.py | 3 +- extensions-builtin/Lora/network_ia3.py | 3 +- extensions-builtin/Lora/network_lokr.py | 3 +- extensions-builtin/Lora/network_lora.py | 72 +++++++++++-------- extensions-builtin/Lora/network_lyco.py | 35 --------- extensions-builtin/Lora/networks.py | 22 ++++-- 10 files changed, 151 insertions(+), 78 deletions(-) create mode 100644 extensions-builtin/Lora/network_full.py delete mode 100644 extensions-builtin/Lora/network_lyco.py diff --git a/extensions-builtin/Lora/extra_networks_lora.py b/extensions-builtin/Lora/extra_networks_lora.py index 8a6639cf..084c41d0 100644 --- a/extensions-builtin/Lora/extra_networks_lora.py +++ b/extensions-builtin/Lora/extra_networks_lora.py @@ -14,14 +14,28 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): params_list.append(extra_networks.ExtraNetworkParams(items=[additional, shared.opts.extra_networks_default_multiplier])) names = [] - multipliers = [] + te_multipliers = [] + unet_multipliers = [] + dyn_dims = [] for params in params_list: assert params.items - names.append(params.items[0]) - multipliers.append(float(params.items[1]) if len(params.items) > 1 else 1.0) + names.append(params.positional[0]) - networks.load_networks(names, multipliers) + te_multiplier = float(params.positional[1]) if len(params.positional) > 1 else 1.0 + te_multiplier = float(params.named.get("te", te_multiplier)) + + unet_multiplier = float(params.positional[2]) if len(params.positional) > 2 else 1.0 + unet_multiplier = float(params.named.get("unet", unet_multiplier)) + + dyn_dim = int(params.positional[3]) if len(params.positional) > 3 else None + dyn_dim = int(params.named["dyn"]) if "dyn" in params.named else dyn_dim + + te_multipliers.append(te_multiplier) + unet_multipliers.append(unet_multiplier) + dyn_dims.append(dyn_dim) + + networks.load_networks(names, te_multipliers, unet_multipliers, dyn_dims) if shared.opts.lora_add_hashes_to_infotext: network_hashes = [] diff --git a/extensions-builtin/Lora/lyco_helpers.py b/extensions-builtin/Lora/lyco_helpers.py index 9ea499fb..279b34bc 100644 --- a/extensions-builtin/Lora/lyco_helpers.py +++ b/extensions-builtin/Lora/lyco_helpers.py @@ -13,3 +13,9 @@ def rebuild_conventional(up, down, shape, dyn_dim=None): up = up[:, :dyn_dim] down = down[:dyn_dim, :] return (up @ down).reshape(shape) + + +def rebuild_cp_decomposition(up, down, mid): + up = up.reshape(up.size(0), -1) + down = down.reshape(down.size(0), -1) + return torch.einsum('n m k l, i n, m j -> i j k l', mid, up, down) diff --git a/extensions-builtin/Lora/network.py b/extensions-builtin/Lora/network.py index 4ac63722..fe42dbdd 100644 --- a/extensions-builtin/Lora/network.py +++ b/extensions-builtin/Lora/network.py @@ -68,7 +68,9 @@ class Network: # LoraModule def __init__(self, name, network_on_disk: NetworkOnDisk): self.name = name self.network_on_disk = network_on_disk - self.multiplier = 1.0 + self.te_multiplier = 1.0 + self.unet_multiplier = 1.0 + self.dyn_dim = None self.modules = {} self.mtime = None @@ -88,6 +90,42 @@ class NetworkModule: self.sd_key = weights.sd_key self.sd_module = weights.sd_module + if hasattr(self.sd_module, 'weight'): + self.shape = self.sd_module.weight.shape + + self.dim = None + self.bias = weights.w.get("bias") + self.alpha = weights.w["alpha"].item() if "alpha" in weights.w else None + self.scale = weights.w["scale"].item() if "scale" in weights.w else None + + def multiplier(self): + if 'transformer' in self.sd_key[:20]: + return self.network.te_multiplier + else: + return self.network.unet_multiplier + + def calc_scale(self): + if self.scale is not None: + return self.scale + if self.dim is not None and self.alpha is not None: + return self.alpha / self.dim + + return 1.0 + + def finalize_updown(self, updown, orig_weight, output_shape): + if self.bias is not None: + updown = updown.reshape(self.bias.shape) + updown += self.bias.to(orig_weight.device, dtype=orig_weight.dtype) + updown = updown.reshape(output_shape) + + if len(output_shape) == 4: + updown = updown.reshape(output_shape) + + if orig_weight.size().numel() == updown.size().numel(): + updown = updown.reshape(orig_weight.shape) + + return updown * self.calc_scale() * self.multiplier() + def calc_updown(self, target): raise NotImplementedError() diff --git a/extensions-builtin/Lora/network_full.py b/extensions-builtin/Lora/network_full.py new file mode 100644 index 00000000..f0d8a6e0 --- /dev/null +++ b/extensions-builtin/Lora/network_full.py @@ -0,0 +1,23 @@ +import lyco_helpers +import network + + +class ModuleTypeFull(network.ModuleType): + def create_module(self, net: network.Network, weights: network.NetworkWeights): + if all(x in weights.w for x in ["diff"]): + return NetworkModuleFull(net, weights) + + return None + + +class NetworkModuleFull(network.NetworkModule): + def __init__(self, net: network.Network, weights: network.NetworkWeights): + super().__init__(net, weights) + + self.weight = weights.w.get("diff") + + def calc_updown(self, orig_weight): + output_shape = self.weight.shape + updown = self.weight.to(orig_weight.device, dtype=orig_weight.dtype) + + return self.finalize_updown(updown, orig_weight, output_shape) diff --git a/extensions-builtin/Lora/network_hada.py b/extensions-builtin/Lora/network_hada.py index 799bb3bc..5fcb0695 100644 --- a/extensions-builtin/Lora/network_hada.py +++ b/extensions-builtin/Lora/network_hada.py @@ -1,6 +1,5 @@ import lyco_helpers import network -import network_lyco class ModuleTypeHada(network.ModuleType): @@ -11,7 +10,7 @@ class ModuleTypeHada(network.ModuleType): return None -class NetworkModuleHada(network_lyco.NetworkModuleLyco): +class NetworkModuleHada(network.NetworkModule): def __init__(self, net: network.Network, weights: network.NetworkWeights): super().__init__(net, weights) diff --git a/extensions-builtin/Lora/network_ia3.py b/extensions-builtin/Lora/network_ia3.py index d8806da0..7edc4249 100644 --- a/extensions-builtin/Lora/network_ia3.py +++ b/extensions-builtin/Lora/network_ia3.py @@ -1,5 +1,4 @@ import network -import network_lyco class ModuleTypeIa3(network.ModuleType): @@ -10,7 +9,7 @@ class ModuleTypeIa3(network.ModuleType): return None -class NetworkModuleIa3(network_lyco.NetworkModuleLyco): +class NetworkModuleIa3(network.NetworkModule): def __init__(self, net: network.Network, weights: network.NetworkWeights): super().__init__(net, weights) diff --git a/extensions-builtin/Lora/network_lokr.py b/extensions-builtin/Lora/network_lokr.py index f1731924..920062e2 100644 --- a/extensions-builtin/Lora/network_lokr.py +++ b/extensions-builtin/Lora/network_lokr.py @@ -2,7 +2,6 @@ import torch import lyco_helpers import network -import network_lyco class ModuleTypeLokr(network.ModuleType): @@ -22,7 +21,7 @@ def make_kron(orig_shape, w1, w2): return torch.kron(w1, w2).reshape(orig_shape) -class NetworkModuleLokr(network_lyco.NetworkModuleLyco): +class NetworkModuleLokr(network.NetworkModule): def __init__(self, net: network.Network, weights: network.NetworkWeights): super().__init__(net, weights) diff --git a/extensions-builtin/Lora/network_lora.py b/extensions-builtin/Lora/network_lora.py index b2d96537..26c0a72c 100644 --- a/extensions-builtin/Lora/network_lora.py +++ b/extensions-builtin/Lora/network_lora.py @@ -1,5 +1,6 @@ import torch +import lyco_helpers import network from modules import devices @@ -16,29 +17,42 @@ class NetworkModuleLora(network.NetworkModule): def __init__(self, net: network.Network, weights: network.NetworkWeights): super().__init__(net, weights) - self.up = self.create_module(weights.w["lora_up.weight"]) - self.down = self.create_module(weights.w["lora_down.weight"]) - self.alpha = weights.w["alpha"] if "alpha" in weights.w else None + self.up_model = self.create_module(weights.w, "lora_up.weight") + self.down_model = self.create_module(weights.w, "lora_down.weight") + self.mid_model = self.create_module(weights.w, "lora_mid.weight", none_ok=True) + + self.dim = weights.w["lora_down.weight"].shape[0] + + def create_module(self, weights, key, none_ok=False): + weight = weights.get(key) - def create_module(self, weight, none_ok=False): if weight is None and none_ok: return None - if type(self.sd_module) == torch.nn.Linear: + is_linear = type(self.sd_module) in [torch.nn.Linear, torch.nn.modules.linear.NonDynamicallyQuantizableLinear, torch.nn.MultiheadAttention] + is_conv = type(self.sd_module) in [torch.nn.Conv2d] + + if is_linear: + weight = weight.reshape(weight.shape[0], -1) module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) - elif type(self.sd_module) == torch.nn.modules.linear.NonDynamicallyQuantizableLinear: - module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) - elif type(self.sd_module) == torch.nn.MultiheadAttention: - module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False) - elif type(self.sd_module) == torch.nn.Conv2d and weight.shape[2:] == (1, 1): + elif is_conv and key == "lora_down.weight" or key == "dyn_up": + if len(weight.shape) == 2: + weight = weight.reshape(weight.shape[0], -1, 1, 1) + + if weight.shape[2] != 1 or weight.shape[3] != 1: + module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], self.sd_module.kernel_size, self.sd_module.stride, self.sd_module.padding, bias=False) + else: + module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (1, 1), bias=False) + elif is_conv and key == "lora_mid.weight": + module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], self.sd_module.kernel_size, self.sd_module.stride, self.sd_module.padding, bias=False) + elif is_conv and key == "lora_up.weight" or key == "dyn_down": module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (1, 1), bias=False) - elif type(self.sd_module) == torch.nn.Conv2d and weight.shape[2:] == (3, 3): - module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (3, 3), bias=False) else: - print(f'Network layer {self.network_key} matched a layer with unsupported type: {type(self.sd_module).__name__}') - return None + raise AssertionError(f'Lora layer {self.network_key} matched a layer with unsupported type: {type(self.sd_module).__name__}') with torch.no_grad(): + if weight.shape != module.weight.shape: + weight = weight.reshape(module.weight.shape) module.weight.copy_(weight) module.to(device=devices.cpu, dtype=devices.dtype) @@ -46,25 +60,27 @@ class NetworkModuleLora(network.NetworkModule): return module - def calc_updown(self, target): - up = self.up.weight.to(target.device, dtype=target.dtype) - down = self.down.weight.to(target.device, dtype=target.dtype) + def calc_updown(self, orig_weight): + up = self.up_model.weight.to(orig_weight.device, dtype=orig_weight.dtype) + down = self.down_model.weight.to(orig_weight.device, dtype=orig_weight.dtype) - if up.shape[2:] == (1, 1) and down.shape[2:] == (1, 1): - updown = (up.squeeze(2).squeeze(2) @ down.squeeze(2).squeeze(2)).unsqueeze(2).unsqueeze(3) - elif up.shape[2:] == (3, 3) or down.shape[2:] == (3, 3): - updown = torch.nn.functional.conv2d(down.permute(1, 0, 2, 3), up).permute(1, 0, 2, 3) + output_shape = [up.size(0), down.size(1)] + if self.mid_model is not None: + # cp-decomposition + mid = self.mid_model.weight.to(orig_weight.device, dtype=orig_weight.dtype) + updown = lyco_helpers.rebuild_cp_decomposition(up, down, mid) + output_shape += mid.shape[2:] else: - updown = up @ down + if len(down.shape) == 4: + output_shape += down.shape[2:] + updown = lyco_helpers.rebuild_conventional(up, down, output_shape, self.network.dyn_dim) - updown = updown * self.network.multiplier * (self.alpha / self.up.weight.shape[1] if self.alpha else 1.0) - - return updown + return self.finalize_updown(updown, orig_weight, output_shape) def forward(self, x, y): - self.up.to(device=devices.device) - self.down.to(device=devices.device) + self.up_model.to(device=devices.device) + self.down_model.to(device=devices.device) - return y + self.up(self.down(x)) * self.network.multiplier * (self.alpha / self.up.weight.shape[1] if self.alpha else 1.0) + return y + self.up_model(self.down_model(x)) * self.multiplier() * self.calc_scale() diff --git a/extensions-builtin/Lora/network_lyco.py b/extensions-builtin/Lora/network_lyco.py deleted file mode 100644 index fc135314..00000000 --- a/extensions-builtin/Lora/network_lyco.py +++ /dev/null @@ -1,35 +0,0 @@ -import network - - -class NetworkModuleLyco(network.NetworkModule): - def __init__(self, net: network.Network, weights: network.NetworkWeights): - super().__init__(net, weights) - - if hasattr(self.sd_module, 'weight'): - self.shape = self.sd_module.weight.shape - - self.dim = None - self.bias = weights.w.get("bias") - self.alpha = weights.w["alpha"].item() if "alpha" in weights.w else None - self.scale = weights.w["scale"].item() if "scale" in weights.w else None - - def finalize_updown(self, updown, orig_weight, output_shape): - if self.bias is not None: - updown = updown.reshape(self.bias.shape) - updown += self.bias.to(orig_weight.device, dtype=orig_weight.dtype) - updown = updown.reshape(output_shape) - - if len(output_shape) == 4: - updown = updown.reshape(output_shape) - - if orig_weight.size().numel() == updown.size().numel(): - updown = updown.reshape(orig_weight.shape) - - scale = ( - self.scale if self.scale is not None - else self.alpha / self.dim if self.dim is not None and self.alpha is not None - else 1.0 - ) - - return updown * scale * self.network.multiplier - diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index 1b358561..401430e8 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -6,6 +6,7 @@ import network_lora import network_hada import network_ia3 import network_lokr +import network_full import torch from typing import Union @@ -17,6 +18,7 @@ module_types = [ network_hada.ModuleTypeHada(), network_ia3.ModuleTypeIa3(), network_lokr.ModuleTypeLokr(), + network_full.ModuleTypeFull(), ] @@ -52,6 +54,15 @@ def convert_diffusers_name_to_compvis(key, is_sd2): m = [] + if match(m, r"lora_unet_conv_in(.*)"): + return f'diffusion_model_input_blocks_0_0{m[0]}' + + if match(m, r"lora_unet_conv_out(.*)"): + return f'diffusion_model_out_2{m[0]}' + + if match(m, r"lora_unet_time_embedding_linear_(\d+)(.*)"): + return f"diffusion_model_time_embed_{m[0] * 2 - 2}{m[1]}" + if match(m, r"lora_unet_down_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"): suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3]) return f"diffusion_model_input_blocks_{1 + m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}" @@ -179,7 +190,7 @@ def load_network(name, network_on_disk): return net -def load_networks(names, multipliers=None): +def load_networks(names, te_multipliers=None, unet_multipliers=None, dyn_dims=None): already_loaded = {} for net in loaded_networks: @@ -218,7 +229,9 @@ def load_networks(names, multipliers=None): print(f"Couldn't find network with name {name}") continue - net.multiplier = multipliers[i] if multipliers else 1.0 + net.te_multiplier = te_multipliers[i] if te_multipliers else 1.0 + net.unet_multiplier = unet_multipliers[i] if unet_multipliers else 1.0 + net.dyn_dim = dyn_dims[i] if dyn_dims else 1.0 loaded_networks.append(net) if failed_to_load_networks: @@ -250,7 +263,7 @@ def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn return current_names = getattr(self, "network_current_names", ()) - wanted_names = tuple((x.name, x.multiplier) for x in loaded_networks) + wanted_names = tuple((x.name, x.te_multiplier, x.unet_multiplier, x.dyn_dim) for x in loaded_networks) weights_backup = getattr(self, "network_weights_backup", None) if weights_backup is None: @@ -288,9 +301,10 @@ def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn 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 += module_out.calc_updown(self.out_proj.weight) + self.out_proj.weight += updown_out continue if module is None: From 2e07a8ae6b1d92838b3a8a0f6eaf5fcf4a92d48f Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 09:05:18 +0300 Subject: [PATCH 011/147] some backwards compatibility linter --- extensions-builtin/Lora/lora.py | 9 +++++++++ extensions-builtin/Lora/network_full.py | 1 - extensions-builtin/Lora/scripts/lora_script.py | 1 + 3 files changed, 10 insertions(+), 1 deletion(-) create mode 100644 extensions-builtin/Lora/lora.py diff --git a/extensions-builtin/Lora/lora.py b/extensions-builtin/Lora/lora.py new file mode 100644 index 00000000..9365aa74 --- /dev/null +++ b/extensions-builtin/Lora/lora.py @@ -0,0 +1,9 @@ +import networks + +list_available_loras = networks.list_available_networks + +available_loras = networks.available_networks +available_lora_aliases = networks.available_network_aliases +available_lora_hash_lookup = networks.available_network_hash_lookup +forbidden_lora_aliases = networks.forbidden_network_aliases +loaded_loras = networks.loaded_networks diff --git a/extensions-builtin/Lora/network_full.py b/extensions-builtin/Lora/network_full.py index f0d8a6e0..109b4c2c 100644 --- a/extensions-builtin/Lora/network_full.py +++ b/extensions-builtin/Lora/network_full.py @@ -1,4 +1,3 @@ -import lyco_helpers import network diff --git a/extensions-builtin/Lora/scripts/lora_script.py b/extensions-builtin/Lora/scripts/lora_script.py index 81e6572a..4c75821e 100644 --- a/extensions-builtin/Lora/scripts/lora_script.py +++ b/extensions-builtin/Lora/scripts/lora_script.py @@ -6,6 +6,7 @@ from fastapi import FastAPI import network import networks +import lora # noqa:F401 import extra_networks_lora import ui_extra_networks_lora from modules import script_callbacks, ui_extra_networks, extra_networks, shared From 35510f7529dc05437a82496187ef06b852be9ab1 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 10:06:02 +0300 Subject: [PATCH 012/147] add alias to lyco network read networks from LyCORIS dir if it exists add credits --- README.md | 1 + extensions-builtin/Lora/networks.py | 3 ++- extensions-builtin/Lora/scripts/lora_script.py | 5 ++++- modules/extra_networks.py | 16 ++++++++++++++-- 4 files changed, 21 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index e6d8e4bd..b796d150 100644 --- a/README.md +++ b/README.md @@ -168,5 +168,6 @@ Licenses for borrowed code can be found in `Settings -> Licenses` screen, and al - Security advice - RyotaK - UniPC sampler - Wenliang Zhao - https://github.com/wl-zhao/UniPC - TAESD - Ollin Boer Bohan - https://github.com/madebyollin/taesd +- LyCORIS - KohakuBlueleaf - Initial Gradio script - posted on 4chan by an Anonymous user. Thank you Anonymous user. - (You) diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index 401430e8..7b4c0312 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -11,7 +11,7 @@ import network_full import torch from typing import Union -from modules import shared, devices, sd_models, errors, scripts, sd_hijack +from modules import shared, devices, sd_models, errors, scripts, sd_hijack, paths module_types = [ network_lora.ModuleTypeLora(), @@ -399,6 +399,7 @@ def list_available_networks(): os.makedirs(shared.cmd_opts.lora_dir, exist_ok=True) candidates = list(shared.walk_files(shared.cmd_opts.lora_dir, allowed_extensions=[".pt", ".ckpt", ".safetensors"])) + candidates += list(shared.walk_files(os.path.join(paths.models_path, "LyCORIS"), allowed_extensions=[".pt", ".ckpt", ".safetensors"])) for filename in candidates: if os.path.isdir(filename): continue diff --git a/extensions-builtin/Lora/scripts/lora_script.py b/extensions-builtin/Lora/scripts/lora_script.py index 4c75821e..f478f718 100644 --- a/extensions-builtin/Lora/scripts/lora_script.py +++ b/extensions-builtin/Lora/scripts/lora_script.py @@ -22,7 +22,10 @@ def unload(): def before_ui(): ui_extra_networks.register_page(ui_extra_networks_lora.ExtraNetworksPageLora()) - extra_networks.register_extra_network(extra_networks_lora.ExtraNetworkLora()) + + extra_network = extra_networks_lora.ExtraNetworkLora() + extra_networks.register_extra_network(extra_network) + extra_networks.register_extra_network_alias(extra_network, "lyco") if not hasattr(torch.nn, 'Linear_forward_before_network'): diff --git a/modules/extra_networks.py b/modules/extra_networks.py index 41799b0a..6ae07e91 100644 --- a/modules/extra_networks.py +++ b/modules/extra_networks.py @@ -4,16 +4,22 @@ from collections import defaultdict from modules import errors extra_network_registry = {} +extra_network_aliases = {} def initialize(): extra_network_registry.clear() + extra_network_aliases.clear() def register_extra_network(extra_network): extra_network_registry[extra_network.name] = extra_network +def register_extra_network_alias(extra_network, alias): + extra_network_aliases[alias] = extra_network + + def register_default_extra_networks(): from modules.extra_networks_hypernet import ExtraNetworkHypernet register_extra_network(ExtraNetworkHypernet()) @@ -82,20 +88,26 @@ def activate(p, extra_network_data): """call activate for extra networks in extra_network_data in specified order, then call activate for all remaining registered networks with an empty argument list""" + activated = [] + for extra_network_name, extra_network_args in extra_network_data.items(): extra_network = extra_network_registry.get(extra_network_name, None) + + if extra_network is None: + extra_network = extra_network_aliases.get(extra_network_name, None) + if extra_network is None: print(f"Skipping unknown extra network: {extra_network_name}") continue try: extra_network.activate(p, extra_network_args) + activated.append(extra_network) except Exception as e: errors.display(e, f"activating extra network {extra_network_name} with arguments {extra_network_args}") for extra_network_name, extra_network in extra_network_registry.items(): - args = extra_network_data.get(extra_network_name, None) - if args is not None: + if extra_network in activated: continue try: From 699108bfbb05c2a7d2ee4a2c7abcfaa0a244d8ea Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 18:56:14 +0300 Subject: [PATCH 013/147] hide cards for networks of incompatible stable diffusion version in Lora extra networks interface --- extensions-builtin/Lora/network.py | 20 +++++++++++ .../Lora/scripts/lora_script.py | 2 ++ .../Lora/ui_edit_user_metadata.py | 20 ++++++++--- .../Lora/ui_extra_networks_lora.py | 34 ++++++++++++++++--- html/extra-networks-card.html | 2 +- javascript/extraNetworks.js | 2 +- modules/sd_models.py | 3 ++ modules/ui_extra_networks.py | 3 +- modules/ui_extra_networks_user_metadata.py | 7 +++- style.css | 6 +++- 10 files changed, 84 insertions(+), 15 deletions(-) diff --git a/extensions-builtin/Lora/network.py b/extensions-builtin/Lora/network.py index fe42dbdd..8ecfa29a 100644 --- a/extensions-builtin/Lora/network.py +++ b/extensions-builtin/Lora/network.py @@ -1,5 +1,6 @@ import os from collections import namedtuple +import enum from modules import sd_models, cache, errors, hashes, shared @@ -8,6 +9,13 @@ NetworkWeights = namedtuple('NetworkWeights', ['network_key', 'sd_key', 'w', 'sd metadata_tags_order = {"ss_sd_model_name": 1, "ss_resolution": 2, "ss_clip_skip": 3, "ss_num_train_images": 10, "ss_tag_frequency": 20} +class SdVersion(enum.Enum): + Unknown = 1 + SD1 = 2 + SD2 = 3 + SDXL = 4 + + class NetworkOnDisk: def __init__(self, name, filename): self.name = name @@ -44,6 +52,18 @@ class NetworkOnDisk: '' ) + self.sd_version = self.detect_version() + + def detect_version(self): + if str(self.metadata.get('ss_base_model_version', "")).startswith("sdxl_"): + return SdVersion.SDXL + elif str(self.metadata.get('ss_v2', "")) == "True": + return SdVersion.SD2 + elif len(self.metadata): + return SdVersion.SD1 + + return SdVersion.Unknown + def set_hash(self, v): self.hash = v self.shorthash = self.hash[0:12] diff --git a/extensions-builtin/Lora/scripts/lora_script.py b/extensions-builtin/Lora/scripts/lora_script.py index f478f718..cd28afc9 100644 --- a/extensions-builtin/Lora/scripts/lora_script.py +++ b/extensions-builtin/Lora/scripts/lora_script.py @@ -63,6 +63,8 @@ shared.options_templates.update(shared.options_section(('extra_networks', "Extra "sd_lora": shared.OptionInfo("None", "Add network to prompt", gr.Dropdown, lambda: {"choices": ["None", *networks.available_networks]}, refresh=networks.list_available_networks), "lora_preferred_name": shared.OptionInfo("Alias from file", "When adding to prompt, refer to Lora by", gr.Radio, {"choices": ["Alias from file", "Filename"]}), "lora_add_hashes_to_infotext": shared.OptionInfo(True, "Add Lora hashes to infotext"), + "lora_show_all": shared.OptionInfo(False, "Always show all networks on the Lora page").info("otherwise, those detected as for incompatible version of Stable Diffusion will be hidden"), + "lora_hide_unknown_for_versions": shared.OptionInfo([], "Hide networks of unknown versions for model versions", gr.CheckboxGroup, {"choices": ["SD1", "SD2", "SDXL"]}), })) diff --git a/extensions-builtin/Lora/ui_edit_user_metadata.py b/extensions-builtin/Lora/ui_edit_user_metadata.py index 354a1d68..c8730443 100644 --- a/extensions-builtin/Lora/ui_edit_user_metadata.py +++ b/extensions-builtin/Lora/ui_edit_user_metadata.py @@ -46,14 +46,17 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) def __init__(self, ui, tabname, page): super().__init__(ui, tabname, page) + self.select_sd_version = None + self.taginfo = None self.edit_activation_text = None self.slider_preferred_weight = None self.edit_notes = None - def save_lora_user_metadata(self, name, desc, activation_text, preferred_weight, notes): + def save_lora_user_metadata(self, name, desc, sd_version, activation_text, preferred_weight, notes): user_metadata = self.get_user_metadata(name) user_metadata["description"] = desc + user_metadata["sd version"] = sd_version user_metadata["activation text"] = activation_text user_metadata["preferred weight"] = preferred_weight user_metadata["notes"] = notes @@ -112,11 +115,11 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) gradio_tags = [(tag, str(count)) for tag, count in tags[0:24]] return [ - *values[0:4], + *values[0:5], + item.get("sd_version", "Unknown"), gr.HighlightedText.update(value=gradio_tags, visible=True if tags else False), user_metadata.get('activation text', ''), float(user_metadata.get('preferred weight', 0.0)), - user_metadata.get('notes', ''), gr.update(visible=True if tags else False), gr.update(value=self.generate_random_prompt_from_tags(tags), visible=True if tags else False), ] @@ -141,10 +144,15 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) return ", ".join(sorted(res)) + def create_extra_default_items_in_left_column(self): + + # this would be a lot better as gr.Radio but I can't make it work + self.select_sd_version = gr.Dropdown(['SD1', 'SD2', 'SDXL', 'Unknown'], value='Unknown', label='Stable Diffusion version', interactive=True) + def create_editor(self): self.create_default_editor_elems() - self.taginfo = gr.HighlightedText(label="Tags") + self.taginfo = gr.HighlightedText(label="Training dataset tags") self.edit_activation_text = gr.Text(label='Activation text', info="Will be added to prompt along with Lora") self.slider_preferred_weight = gr.Slider(label='Preferred weight', info="Set to 0 to disable", minimum=0.0, maximum=2.0, step=0.01) @@ -178,10 +186,11 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) self.edit_description, self.html_filedata, self.html_preview, + self.edit_notes, + self.select_sd_version, self.taginfo, self.edit_activation_text, self.slider_preferred_weight, - self.edit_notes, row_random_prompt, random_prompt, ] @@ -192,6 +201,7 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) edited_components = [ self.edit_description, + self.select_sd_version, self.edit_activation_text, self.slider_preferred_weight, self.edit_notes, diff --git a/extensions-builtin/Lora/ui_extra_networks_lora.py b/extensions-builtin/Lora/ui_extra_networks_lora.py index b6171a26..4b32098b 100644 --- a/extensions-builtin/Lora/ui_extra_networks_lora.py +++ b/extensions-builtin/Lora/ui_extra_networks_lora.py @@ -1,7 +1,9 @@ import os + +import network import networks -from modules import shared, ui_extra_networks +from modules import shared, ui_extra_networks, paths from modules.ui_extra_networks import quote_js from ui_edit_user_metadata import LoraUserMetadataEditor @@ -13,14 +15,13 @@ class ExtraNetworksPageLora(ui_extra_networks.ExtraNetworksPage): def refresh(self): networks.list_available_networks() - def create_item(self, name, index=None): + def create_item(self, name, index=None, enable_filter=True): lora_on_disk = networks.available_networks.get(name) path, ext = os.path.splitext(lora_on_disk.filename) alias = lora_on_disk.get_alias() - # in 1.5 filename changes to be full filename instead of path without extension, and metadata is dict instead of json string item = { "name": name, "filename": lora_on_disk.filename, @@ -30,6 +31,7 @@ class ExtraNetworksPageLora(ui_extra_networks.ExtraNetworksPage): "local_preview": f"{path}.{shared.opts.samples_format}", "metadata": lora_on_disk.metadata, "sort_keys": {'default': index, **self.get_sort_keys(lora_on_disk.filename)}, + "sd_version": lora_on_disk.sd_version.name, } self.read_user_metadata(item) @@ -40,15 +42,37 @@ class ExtraNetworksPageLora(ui_extra_networks.ExtraNetworksPage): if activation_text: item["prompt"] += " + " + quote_js(" " + activation_text) + sd_version = item["user_metadata"].get("sd version") + if sd_version in network.SdVersion.__members__: + item["sd_version"] = sd_version + sd_version = network.SdVersion[sd_version] + else: + sd_version = lora_on_disk.sd_version + + if shared.opts.lora_show_all or not enable_filter: + pass + elif sd_version == network.SdVersion.Unknown: + model_version = network.SdVersion.SDXL if shared.sd_model.is_sdxl else network.SdVersion.SD2 if shared.sd_model.is_sd2 else network.SdVersion.SD1 + if model_version.name in shared.opts.lora_hide_unknown_for_versions: + return None + elif shared.sd_model.is_sdxl and sd_version != network.SdVersion.SDXL: + return None + elif shared.sd_model.is_sd2 and sd_version != network.SdVersion.SD2: + return None + elif shared.sd_model.is_sd1 and sd_version != network.SdVersion.SD1: + return None + return item def list_items(self): for index, name in enumerate(networks.available_networks): item = self.create_item(name, index) - yield item + + if item is not None: + yield item def allowed_directories_for_previews(self): - return [shared.cmd_opts.lora_dir] + return [shared.cmd_opts.lora_dir, os.path.join(paths.models_path, "LyCORIS")] def create_user_metadata_editor(self, ui, tabname): return LoraUserMetadataEditor(ui, tabname, self) diff --git a/html/extra-networks-card.html b/html/extra-networks-card.html index eb8b1a67..39674666 100644 --- a/html/extra-networks-card.html +++ b/html/extra-networks-card.html @@ -1,8 +1,8 @@
{background_image}
- {edit_button} {metadata_button} + {edit_button}
diff --git a/javascript/extraNetworks.js b/javascript/extraNetworks.js index e453094a..5582a6e5 100644 --- a/javascript/extraNetworks.js +++ b/javascript/extraNetworks.js @@ -213,7 +213,7 @@ function popup(contents) { globalPopupInner.classList.add('global-popup-inner'); globalPopup.appendChild(globalPopupInner); - gradioApp().appendChild(globalPopup); + gradioApp().querySelector('.main').appendChild(globalPopup); } globalPopupInner.innerHTML = ''; diff --git a/modules/sd_models.py b/modules/sd_models.py index 729f03d7..4d9382dd 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -290,6 +290,9 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer state_dict = get_checkpoint_state_dict(checkpoint_info, timer) model.is_sdxl = hasattr(model, 'conditioner') + model.is_sd2 = not model.is_sdxl and hasattr(model.cond_stage_model, 'model') + model.is_sd1 = not model.is_sdxl and not model.is_sd2 + if model.is_sdxl: sd_models_xl.extend_sdxl(model) diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index 6c73998f..49612298 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -62,7 +62,8 @@ def get_single_card(page: str = "", tabname: str = "", name: str = ""): page = next(iter([x for x in extra_pages if x.name == page]), None) try: - item = page.create_item(name) + item = page.create_item(name, enable_filter=False) + page.items[name] = item except Exception as e: errors.display(e, "creating item for extra network") item = page.items.get(name) diff --git a/modules/ui_extra_networks_user_metadata.py b/modules/ui_extra_networks_user_metadata.py index 01ff4e4b..63d4b503 100644 --- a/modules/ui_extra_networks_user_metadata.py +++ b/modules/ui_extra_networks_user_metadata.py @@ -42,6 +42,9 @@ class UserMetadataEditor: return user_metadata + def create_extra_default_items_in_left_column(self): + pass + def create_default_editor_elems(self): with gr.Row(): with gr.Column(scale=2): @@ -49,6 +52,8 @@ class UserMetadataEditor: self.edit_description = gr.Textbox(label="Description", lines=4) self.html_filedata = gr.HTML() + self.create_extra_default_items_in_left_column() + with gr.Column(scale=1, min_width=0): self.html_preview = gr.HTML() @@ -111,7 +116,7 @@ class UserMetadataEditor: table = '' + "".join(f"" for name, value in params) + '
{name}{value}
' - return html.escape(name), user_metadata.get('description', ''), table, self.get_card_html(name), user_metadata.get('notes', ''), + return html.escape(name), user_metadata.get('description', ''), table, self.get_card_html(name), user_metadata.get('notes', '') def write_user_metadata(self, name, metadata): item = self.page.items.get(name, {}) diff --git a/style.css b/style.css index 8a66c3d2..e249cfd3 100644 --- a/style.css +++ b/style.css @@ -841,7 +841,7 @@ footer { .extra-network-cards .card .card-button { text-shadow: 2px 2px 3px black; - padding: 0.25em; + padding: 0.25em 0.1em; font-size: 200%; width: 1.5em; } @@ -957,6 +957,10 @@ div.block.gradio-box.edit-user-metadata { text-align: left; } +.edit-user-metadata .file-metadata th, .edit-user-metadata .file-metadata td{ + padding: 0.3em 1em; +} + .edit-user-metadata .wrap.translucent{ background: var(--body-background-fill); } From a99d5708e6d603e8f7cfd1b8c6595f8026219ba0 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 17 Jul 2023 20:10:24 +0300 Subject: [PATCH 014/147] skip installing packages with pip if theyare already installed record time it took to launch --- modules/launch_utils.py | 46 ++++++++++++++++++++++++++++++++++++++- requirements_versions.txt | 4 ++-- webui.py | 9 ++++---- 3 files changed, 52 insertions(+), 7 deletions(-) diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 434facbc..03552bc2 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -1,4 +1,5 @@ # this scripts installs necessary requirements and launches main program in webui.py +import re import subprocess import os import sys @@ -9,6 +10,9 @@ from functools import lru_cache from modules import cmd_args, errors from modules.paths_internal import script_path, extensions_dir +from modules import timer + +timer.startup_timer.record("start") args, _ = cmd_args.parser.parse_known_args() @@ -226,6 +230,44 @@ def run_extensions_installers(settings_file): run_extension_installer(os.path.join(extensions_dir, dirname_extension)) +re_requirement = re.compile(r"\s*([-_a-zA-Z0-9]+)\s*(?:==\s*([-+_.a-zA-Z0-9]+))?\s*") + + +def requrements_met(requirements_file): + """ + Does a simple parse of a requirements.txt file to determine if all rerqirements in it + are already installed. Returns True if so, False if not installed or parsing fails. + """ + + import importlib.metadata + import packaging.version + + with open(requirements_file, "r", encoding="utf8") as file: + for line in file: + if line.strip() == "": + continue + + m = re.match(re_requirement, line) + if m is None: + return False + + package = m.group(1).strip() + version_required = (m.group(2) or "").strip() + + if version_required == "": + continue + + try: + version_installed = importlib.metadata.version(package) + except Exception: + return False + + if packaging.version.parse(version_required) != packaging.version.parse(version_installed): + return False + + return True + + def prepare_environment(): torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu118") torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.0.1 torchvision==0.15.2 --extra-index-url {torch_index_url}") @@ -311,7 +353,9 @@ def prepare_environment(): if not os.path.isfile(requirements_file): requirements_file = os.path.join(script_path, requirements_file) - run_pip(f"install -r \"{requirements_file}\"", "requirements") + + if not requrements_met(requirements_file): + run_pip(f"install -r \"{requirements_file}\"", "requirements") run_extensions_installers(settings_file=args.ui_settings_file) diff --git a/requirements_versions.txt b/requirements_versions.txt index b826bf43..d07ab456 100644 --- a/requirements_versions.txt +++ b/requirements_versions.txt @@ -8,7 +8,7 @@ einops==0.4.1 fastapi==0.94.0 gfpgan==1.3.8 gradio==3.32.0 -httpcore<=0.15 +httpcore==0.15 inflection==0.5.1 jsonmerge==1.8.0 kornia==0.6.7 @@ -17,7 +17,7 @@ numpy==1.23.5 omegaconf==2.2.3 open-clip-torch==2.20.0 piexif==1.1.3 -psutil~=5.9.5 +psutil==5.9.5 pytorch_lightning==1.9.4 realesrgan==0.3.0 resize-right==0.0.2 diff --git a/webui.py b/webui.py index 34c2fd18..2aafc09f 100644 --- a/webui.py +++ b/webui.py @@ -31,21 +31,22 @@ if log_level: logging.getLogger("torch.distributed.nn").setLevel(logging.ERROR) # sshh... logging.getLogger("xformers").addFilter(lambda record: 'A matching Triton is not available' not in record.getMessage()) -from modules import paths, timer, import_hook, errors, devices # noqa: F401 - +from modules import timer startup_timer = timer.startup_timer +startup_timer.record("launcher") import torch import pytorch_lightning # noqa: F401 # pytorch_lightning should be imported after torch, but it re-enables warnings on import so import once to disable them warnings.filterwarnings(action="ignore", category=DeprecationWarning, module="pytorch_lightning") warnings.filterwarnings(action="ignore", category=UserWarning, module="torchvision") - - startup_timer.record("import torch") import gradio # noqa: F401 startup_timer.record("import gradio") +from modules import paths, timer, import_hook, errors, devices # noqa: F401 +startup_timer.record("setup paths") + import ldm.modules.encoders.modules # noqa: F401 startup_timer.record("import ldm") From 17e14ed2d9451859325d275ccc6cdf51fc85a56d Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 18 Jul 2023 10:23:41 +0800 Subject: [PATCH 015/147] Fix wrong key name in lokr module --- extensions-builtin/Lora/network_lokr.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/extensions-builtin/Lora/network_lokr.py b/extensions-builtin/Lora/network_lokr.py index 920062e2..3a94f3e9 100644 --- a/extensions-builtin/Lora/network_lokr.py +++ b/extensions-builtin/Lora/network_lokr.py @@ -6,8 +6,8 @@ import network class ModuleTypeLokr(network.ModuleType): def create_module(self, net: network.Network, weights: network.NetworkWeights): - has_1 = "lokr_w1" in weights.w or ("lokr_w1a" in weights.w and "lokr_w1b" in weights.w) - has_2 = "lokr_w2" in weights.w or ("lokr_w2a" in weights.w and "lokr_w2b" in weights.w) + has_1 = "lokr_w1" in weights.w or ("lokr_w1_a" in weights.w and "lokr_w1_b" in weights.w) + has_2 = "lokr_w2" in weights.w or ("lokr_w2_a" in weights.w and "lokr_w2_b" in weights.w) if has_1 and has_2: return NetworkModuleLokr(net, weights) From 3d31caf4a53c4bb4469b72790b459eba7b251da9 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 18 Jul 2023 10:45:42 +0800 Subject: [PATCH 016/147] use "is not None" for Tensor --- extensions-builtin/Lora/network_lokr.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/extensions-builtin/Lora/network_lokr.py b/extensions-builtin/Lora/network_lokr.py index 3a94f3e9..340acdab 100644 --- a/extensions-builtin/Lora/network_lokr.py +++ b/extensions-builtin/Lora/network_lokr.py @@ -28,11 +28,11 @@ class NetworkModuleLokr(network.NetworkModule): self.w1 = weights.w.get("lokr_w1") self.w1a = weights.w.get("lokr_w1_a") self.w1b = weights.w.get("lokr_w1_b") - self.dim = self.w1b.shape[0] if self.w1b else self.dim + self.dim = self.w1b.shape[0] if self.w1b is not None else self.dim self.w2 = weights.w.get("lokr_w2") self.w2a = weights.w.get("lokr_w2_a") self.w2b = weights.w.get("lokr_w2_b") - self.dim = self.w2b.shape[0] if self.w2b else self.dim + self.dim = self.w2b.shape[0] if self.w2b is not None else self.dim self.t2 = weights.w.get("lokr_t2") def calc_updown(self, orig_weight): From 40a18d38a8fcb88d1c2947a2653b52cd2085536f Mon Sep 17 00:00:00 2001 From: lambertae Date: Tue, 18 Jul 2023 00:32:01 -0400 Subject: [PATCH 017/147] add restart sampler --- modules/sd_samplers_kdiffusion.py | 70 ++++++++++++++++++++++++++++++- 1 file changed, 68 insertions(+), 2 deletions(-) diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index 71581b76..c63b677c 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -1,3 +1,5 @@ +# export PIP_CACHE_DIR=/scratch/dengm/cache +# export XDG_CACHE_HOME=/scratch/dengm/cache from collections import deque import torch import inspect @@ -30,12 +32,76 @@ samplers_k_diffusion = [ ('DPM++ 2M Karras', 'sample_dpmpp_2m', ['k_dpmpp_2m_ka'], {'scheduler': 'karras'}), ('DPM++ SDE Karras', 'sample_dpmpp_sde', ['k_dpmpp_sde_ka'], {'scheduler': 'karras', "second_order": True, "brownian_noise": True}), ('DPM++ 2M SDE Karras', 'sample_dpmpp_2m_sde', ['k_dpmpp_2m_sde_ka'], {'scheduler': 'karras', "brownian_noise": True}), + ('Restart (new)', 'restart_sampler', ['restart'], {'scheduler': 'karras', "second_order": True}), ] + +@torch.no_grad() +def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=None, s_noise=1., restart_list = {0.1: [10, 2, 2]}): + """Implements restart sampling in Restart Sampling for Improving Generative Processes (2023)""" + '''Restart_list format: {min_sigma: [ restart_steps, restart_times, max_sigma]}''' + + from tqdm.auto import trange, tqdm + extra_args = {} if extra_args is None else extra_args + s_in = x.new_ones([x.shape[0]]) + step_id = 0 + + from k_diffusion.sampling import to_d, append_zero + + def heun_step(x, old_sigma, new_sigma): + nonlocal step_id + denoised = model(x, old_sigma * s_in, **extra_args) + d = to_d(x, old_sigma, denoised) + if callback is not None: + callback({'x': x, 'i': step_id, 'sigma': new_sigma, 'sigma_hat': old_sigma, 'denoised': denoised}) + dt = new_sigma - old_sigma + if new_sigma == 0: + # Euler method + x = x + d * dt + else: + # Heun's method + x_2 = x + d * dt + denoised_2 = model(x_2, new_sigma * s_in, **extra_args) + d_2 = to_d(x_2, new_sigma, denoised_2) + d_prime = (d + d_2) / 2 + x = x + d_prime * dt + step_id += 1 + return x + # print(sigmas) + temp_list = dict() + for key, value in restart_list.items(): + temp_list[int(torch.argmin(abs(sigmas - key), dim=0))] = value + restart_list = temp_list + + + def get_sigmas_karras(n, sigma_min, sigma_max, rho=7., device='cpu'): + ramp = torch.linspace(0, 1, n).to(device) + min_inv_rho = (sigma_min ** (1 / rho)) + max_inv_rho = (sigma_max ** (1 / rho)) + if isinstance(min_inv_rho, torch.Tensor): + min_inv_rho = min_inv_rho.to(device) + if isinstance(max_inv_rho, torch.Tensor): + max_inv_rho = max_inv_rho.to(device) + sigmas = (max_inv_rho + ramp * (min_inv_rho - max_inv_rho)) ** rho + return append_zero(sigmas).to(device) + + for i in trange(len(sigmas) - 1, disable=disable): + x = heun_step(x, sigmas[i], sigmas[i+1]) + if i + 1 in restart_list: + restart_steps, restart_times, restart_max = restart_list[i + 1] + min_idx = i + 1 + max_idx = int(torch.argmin(abs(sigmas - restart_max), dim=0)) + sigma_restart = get_sigmas_karras(restart_steps, sigmas[min_idx], sigmas[max_idx], device=sigmas.device)[:-1] # remove the zero at the end + for times in range(restart_times): + x = x + torch.randn_like(x) * s_noise * (sigmas[max_idx] ** 2 - sigmas[min_idx] ** 2) ** 0.5 + for (old_sigma, new_sigma) in zip(sigma_restart[:-1], sigma_restart[1:]): + x = heun_step(x, old_sigma, new_sigma) + return x + samplers_data_k_diffusion = [ sd_samplers_common.SamplerData(label, lambda model, funcname=funcname: KDiffusionSampler(funcname, model), aliases, options) for label, funcname, aliases, options in samplers_k_diffusion - if hasattr(k_diffusion.sampling, funcname) + if (hasattr(k_diffusion.sampling, funcname) or funcname == 'restart_sampler') ] sampler_extra_params = { @@ -245,7 +311,7 @@ class KDiffusionSampler: self.model_wrap = denoiser(sd_model, quantize=shared.opts.enable_quantization) self.funcname = funcname - self.func = getattr(k_diffusion.sampling, self.funcname) + self.func = getattr(k_diffusion.sampling, self.funcname) if funcname != "restart_sampler" else restart_sampler self.extra_params = sampler_extra_params.get(funcname, []) self.model_wrap_cfg = CFGDenoiser(self.model_wrap) self.sampler_noises = None From 15a94d6cf7fa075c09362e73c1239692d021c559 Mon Sep 17 00:00:00 2001 From: lambertae Date: Tue, 18 Jul 2023 00:39:26 -0400 Subject: [PATCH 018/147] remove useless header --- modules/sd_samplers_kdiffusion.py | 2 -- 1 file changed, 2 deletions(-) diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index c63b677c..7888d864 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -1,5 +1,3 @@ -# export PIP_CACHE_DIR=/scratch/dengm/cache -# export XDG_CACHE_HOME=/scratch/dengm/cache from collections import deque import torch import inspect From f0e2098f1a533c88396536282c1d6cd7d847a51c Mon Sep 17 00:00:00 2001 From: brkirch Date: Mon, 17 Jul 2023 23:39:38 -0400 Subject: [PATCH 019/147] Add support for `--upcast-sampling` with SD XL --- modules/sd_hijack_unet.py | 8 +++++++- modules/sd_models.py | 2 +- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/modules/sd_hijack_unet.py b/modules/sd_hijack_unet.py index ca1daf45..2101f1a0 100644 --- a/modules/sd_hijack_unet.py +++ b/modules/sd_hijack_unet.py @@ -39,7 +39,10 @@ def apply_model(orig_func, self, x_noisy, t, cond, **kwargs): if isinstance(cond, dict): for y in cond.keys(): - cond[y] = [x.to(devices.dtype_unet) if isinstance(x, torch.Tensor) else x for x in cond[y]] + if isinstance(cond[y], list): + cond[y] = [x.to(devices.dtype_unet) if isinstance(x, torch.Tensor) else x for x in cond[y]] + else: + cond[y] = cond[y].to(devices.dtype_unet) if isinstance(cond[y], torch.Tensor) else cond[y] with devices.autocast(): return orig_func(self, x_noisy.to(devices.dtype_unet), t.to(devices.dtype_unet), cond, **kwargs).float() @@ -77,3 +80,6 @@ first_stage_sub = lambda orig_func, self, x, **kwargs: orig_func(self, x.to(devi CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.decode_first_stage', first_stage_sub, first_stage_cond) CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.encode_first_stage', first_stage_sub, first_stage_cond) CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.get_first_stage_encoding', lambda orig_func, *args, **kwargs: orig_func(*args, **kwargs).float(), first_stage_cond) + +CondFunc('sgm.modules.diffusionmodules.wrappers.OpenAIWrapper.forward', apply_model, unet_needs_upcast) +CondFunc('sgm.modules.diffusionmodules.openaimodel.timestep_embedding', lambda orig_func, timesteps, *args, **kwargs: orig_func(timesteps, *args, **kwargs).to(torch.float32 if timesteps.dtype == torch.int64 else devices.dtype_unet), unet_needs_upcast) diff --git a/modules/sd_models.py b/modules/sd_models.py index 4d9382dd..5813b550 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -326,7 +326,7 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer timer.record("apply half()") - devices.dtype_unet = model.model.diffusion_model.dtype + devices.dtype_unet = torch.float16 if model.is_sdxl and not shared.cmd_opts.no_half else model.model.diffusion_model.dtype devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 model.first_stage_model.to(devices.dtype_vae) From 37e048a7e2356f4caebfd976351112f03856f082 Mon Sep 17 00:00:00 2001 From: lambertae Date: Tue, 18 Jul 2023 00:55:02 -0400 Subject: [PATCH 020/147] fix floating error --- modules/sd_samplers_kdiffusion.py | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index 7888d864..1bb25adf 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -89,11 +89,12 @@ def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=No restart_steps, restart_times, restart_max = restart_list[i + 1] min_idx = i + 1 max_idx = int(torch.argmin(abs(sigmas - restart_max), dim=0)) - sigma_restart = get_sigmas_karras(restart_steps, sigmas[min_idx], sigmas[max_idx], device=sigmas.device)[:-1] # remove the zero at the end - for times in range(restart_times): - x = x + torch.randn_like(x) * s_noise * (sigmas[max_idx] ** 2 - sigmas[min_idx] ** 2) ** 0.5 - for (old_sigma, new_sigma) in zip(sigma_restart[:-1], sigma_restart[1:]): - x = heun_step(x, old_sigma, new_sigma) + if max_idx < min_idx: + sigma_restart = get_sigmas_karras(restart_steps, sigmas[min_idx], sigmas[max_idx], device=sigmas.device)[:-1] # remove the zero at the end + for times in range(restart_times): + x = x + torch.randn_like(x) * s_noise * (sigmas[max_idx] ** 2 - sigmas[min_idx] ** 2) ** 0.5 + for (old_sigma, new_sigma) in zip(sigma_restart[:-1], sigma_restart[1:]): + x = heun_step(x, old_sigma, new_sigma) return x samplers_data_k_diffusion = [ From 7bb0fbed136c6a345b211e09102659fd89362576 Mon Sep 17 00:00:00 2001 From: lambertae Date: Tue, 18 Jul 2023 01:02:04 -0400 Subject: [PATCH 021/147] code styling --- modules/sd_samplers_kdiffusion.py | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/modules/sd_samplers_kdiffusion.py b/modules/sd_samplers_kdiffusion.py index 1bb25adf..db7013f2 100644 --- a/modules/sd_samplers_kdiffusion.py +++ b/modules/sd_samplers_kdiffusion.py @@ -35,17 +35,15 @@ samplers_k_diffusion = [ @torch.no_grad() -def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=None, s_noise=1., restart_list = {0.1: [10, 2, 2]}): +def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=None, s_noise=1.): """Implements restart sampling in Restart Sampling for Improving Generative Processes (2023)""" '''Restart_list format: {min_sigma: [ restart_steps, restart_times, max_sigma]}''' - - from tqdm.auto import trange, tqdm + restart_list = {0.1: [10, 2, 2]} + from tqdm.auto import trange extra_args = {} if extra_args is None else extra_args s_in = x.new_ones([x.shape[0]]) step_id = 0 - from k_diffusion.sampling import to_d, append_zero - def heun_step(x, old_sigma, new_sigma): nonlocal step_id denoised = model(x, old_sigma * s_in, **extra_args) @@ -70,8 +68,6 @@ def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=No for key, value in restart_list.items(): temp_list[int(torch.argmin(abs(sigmas - key), dim=0))] = value restart_list = temp_list - - def get_sigmas_karras(n, sigma_min, sigma_max, rho=7., device='cpu'): ramp = torch.linspace(0, 1, n).to(device) min_inv_rho = (sigma_min ** (1 / rho)) @@ -82,7 +78,6 @@ def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=No max_inv_rho = max_inv_rho.to(device) sigmas = (max_inv_rho + ramp * (min_inv_rho - max_inv_rho)) ** rho return append_zero(sigmas).to(device) - for i in trange(len(sigmas) - 1, disable=disable): x = heun_step(x, sigmas[i], sigmas[i+1]) if i + 1 in restart_list: @@ -91,7 +86,8 @@ def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=No max_idx = int(torch.argmin(abs(sigmas - restart_max), dim=0)) if max_idx < min_idx: sigma_restart = get_sigmas_karras(restart_steps, sigmas[min_idx], sigmas[max_idx], device=sigmas.device)[:-1] # remove the zero at the end - for times in range(restart_times): + while restart_times > 0: + restart_times -= 1 x = x + torch.randn_like(x) * s_noise * (sigmas[max_idx] ** 2 - sigmas[min_idx] ** 2) ** 0.5 for (old_sigma, new_sigma) in zip(sigma_restart[:-1], sigma_restart[1:]): x = heun_step(x, old_sigma, new_sigma) From d6668347c8b85b11b696ac56777cc396e34ee1f9 Mon Sep 17 00:00:00 2001 From: Leon Feng Date: Tue, 18 Jul 2023 04:19:58 -0400 Subject: [PATCH 022/147] remove duplicate --- modules/textual_inversion/logging.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/textual_inversion/logging.py b/modules/textual_inversion/logging.py index 734a4b6f..a822a7a5 100644 --- a/modules/textual_inversion/logging.py +++ b/modules/textual_inversion/logging.py @@ -2,7 +2,7 @@ import datetime import json import os -saved_params_shared = {"model_name", "model_hash", "initial_step", "num_of_dataset_images", "learn_rate", "batch_size", "clip_grad_mode", "clip_grad_value", "gradient_step", "data_root", "log_directory", "training_width", "training_height", "steps", "create_image_every", "template_file", "gradient_step", "latent_sampling_method"} +saved_params_shared = {"model_name", "model_hash", "initial_step", "num_of_dataset_images", "learn_rate", "batch_size", "clip_grad_mode", "clip_grad_value", "data_root", "log_directory", "training_width", "training_height", "steps", "create_image_every", "template_file", "gradient_step", "latent_sampling_method"} saved_params_ti = {"embedding_name", "num_vectors_per_token", "save_embedding_every", "save_image_with_stored_embedding"} saved_params_hypernet = {"hypernetwork_name", "layer_structure", "activation_func", "weight_init", "add_layer_norm", "use_dropout", "save_hypernetwork_every"} saved_params_all = saved_params_shared | saved_params_ti | saved_params_hypernet From 420cc8f68e6aca8a3a0f42ee0e626a6b03712763 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 11:48:40 +0300 Subject: [PATCH 023/147] also make None a valid option for options API for #11854 --- modules/api/models.py | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/modules/api/models.py b/modules/api/models.py index b5683071..b55fa728 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -1,4 +1,6 @@ import inspect +import types + from pydantic import BaseModel, Field, create_model from typing import Any, Optional from typing_extensions import Literal @@ -207,11 +209,14 @@ class PreprocessResponse(BaseModel): fields = {} for key, metadata in opts.data_labels.items(): value = opts.data.get(key) + if key == 'sd_model_checkpoint': + value = None optType = opts.typemap.get(type(metadata.default), type(value)) - if (metadata is not None): - fields.update({key: (Optional[optType], Field( - default=metadata.default ,description=metadata.label))}) + if optType == types.NoneType: + pass + elif metadata is not None: + fields.update({key: (Optional[optType], Field(default=metadata.default, description=metadata.label))}) else: fields.update({key: (Optional[optType], Field())}) From 3c570421d3a2eb24528b5f5bb615dcb0c7717e4a Mon Sep 17 00:00:00 2001 From: wfjsw Date: Tue, 18 Jul 2023 19:00:16 +0800 Subject: [PATCH 024/147] move start timer --- launch.py | 4 +++- modules/api/models.py | 2 +- modules/launch_utils.py | 3 --- 3 files changed, 4 insertions(+), 5 deletions(-) diff --git a/launch.py b/launch.py index b103c8f3..e9667c88 100644 --- a/launch.py +++ b/launch.py @@ -1,4 +1,4 @@ -from modules import launch_utils +from modules import launch_utils, timer args = launch_utils.args @@ -25,6 +25,8 @@ start = launch_utils.start def main(): + timer.startup_timer.record("start") + if not args.skip_prepare_environment: prepare_environment() diff --git a/modules/api/models.py b/modules/api/models.py index b55fa728..96cfe920 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -213,7 +213,7 @@ for key, metadata in opts.data_labels.items(): value = None optType = opts.typemap.get(type(metadata.default), type(value)) - if optType == types.NoneType: + if isinstance(optType, types.NoneType): pass elif metadata is not None: fields.update({key: (Optional[optType], Field(default=metadata.default, description=metadata.label))}) diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 03552bc2..ea995eda 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -10,9 +10,6 @@ from functools import lru_cache from modules import cmd_args, errors from modules.paths_internal import script_path, extensions_dir -from modules import timer - -timer.startup_timer.record("start") args, _ = cmd_args.parser.parse_known_args() From ed82f1c5f1677c85298f4d2c6c030a5551682c71 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 15:55:23 +0300 Subject: [PATCH 025/147] lint --- modules/api/models.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/api/models.py b/modules/api/models.py index b55fa728..96cfe920 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -213,7 +213,7 @@ for key, metadata in opts.data_labels.items(): value = None optType = opts.typemap.get(type(metadata.default), type(value)) - if optType == types.NoneType: + if isinstance(optType, types.NoneType): pass elif metadata is not None: fields.update({key: (Optional[optType], Field(default=metadata.default, description=metadata.label))}) From 4b5a63aa1135757ef9db58b15f5426e758d285d0 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 17:32:46 +0300 Subject: [PATCH 026/147] add a bit more metadata info for the lora user metadata page --- extensions-builtin/Lora/ui_edit_user_metadata.py | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/extensions-builtin/Lora/ui_edit_user_metadata.py b/extensions-builtin/Lora/ui_edit_user_metadata.py index c8730443..2ca997f7 100644 --- a/extensions-builtin/Lora/ui_edit_user_metadata.py +++ b/extensions-builtin/Lora/ui_edit_user_metadata.py @@ -1,3 +1,4 @@ +import datetime import html import random @@ -71,6 +72,7 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) keys = { 'ss_sd_model_name': "Model:", 'ss_clip_skip': "Clip skip:", + 'ss_network_module': "Kohya module:", } for key, label in keys.items(): @@ -78,6 +80,10 @@ class LoraUserMetadataEditor(ui_extra_networks_user_metadata.UserMetadataEditor) if value is not None and str(value) != "None": table.append((label, html.escape(value))) + ss_training_started_at = metadata.get('ss_training_started_at') + if ss_training_started_at: + table.append(("Date trained:", datetime.datetime.utcfromtimestamp(float(ss_training_started_at)).strftime('%Y-%m-%d %H:%M'))) + ss_bucket_info = metadata.get("ss_bucket_info") if ss_bucket_info and "buckets" in ss_bucket_info: resolutions = {} From 66c5f1bb1556a2d86d9f11aeb92f83d4a09832cc Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 17:41:37 +0300 Subject: [PATCH 027/147] return sd_model_checkpoint to None --- modules/shared.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/shared.py b/modules/shared.py index a256d090..6162938a 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -409,7 +409,7 @@ options_templates.update(options_section(('training', "Training"), { })) options_templates.update(options_section(('sd', "Stable Diffusion"), { - "sd_model_checkpoint": OptionInfo("", "Stable Diffusion checkpoint", gr.Dropdown, lambda: {"choices": list_checkpoint_tiles()}, refresh=refresh_checkpoints), + "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("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"), From be16d274f85a59b742f7480cdb79a596c972f598 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 17:44:56 +0300 Subject: [PATCH 028/147] changelog for 1.5.0 --- CHANGELOG.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/CHANGELOG.md b/CHANGELOG.md index 30783d9d..007010da 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,7 @@ * SD XL support * user metadata system for custom networks * extended Lora metadata editor: set activation text, default weight, view tags, training info + * Lora extension rework to include other types of networks (all that were previously handled by LyCORIS extension) * show github stars for extenstions * img2img batch mode can read extra stuff from png info * img2img batch works with subdirectories @@ -11,6 +12,9 @@ * restyle time taken/VRAM display * add textual inversion hashes to infotext * optimization: cache git extension repo information + * move generate button next to the generated picture for mobile clients + * hide cards for networks of incompatible Stable Diffusion version in Lora extra networks interface + * skip installing packages with pip if they all are already installed - startup speedup of about 2 seconds ### Minor: * checkbox to check/uncheck all extensions in the Installed tab @@ -25,6 +29,7 @@ * speedup extra networks listing * added `[none]` filename token. * removed thumbs extra networks view mode (use settings tab to change width/height/scale to get thumbs) + * add always_discard_next_to_last_sigma option to XYZ plot ### Extensions and API: * api endpoints: /sdapi/v1/server-kill, /sdapi/v1/server-restart, /sdapi/v1/server-stop From b270ded268c92950a35a7a326da54496ef4151c8 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 18:10:04 +0300 Subject: [PATCH 029/147] fix the issue with /sdapi/v1/options failing (this time for sure!) fix automated tests downloading CLIP model --- .github/workflows/run_tests.yaml | 1 + modules/api/models.py | 6 ++---- modules/cmd_args.py | 1 + modules/sd_models.py | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) diff --git a/.github/workflows/run_tests.yaml b/.github/workflows/run_tests.yaml index e9370cc0..3dafaf8d 100644 --- a/.github/workflows/run_tests.yaml +++ b/.github/workflows/run_tests.yaml @@ -41,6 +41,7 @@ jobs: --skip-prepare-environment --skip-torch-cuda-test --test-server + --do-not-download-clip --no-half --disable-opt-split-attention --use-cpu all diff --git a/modules/api/models.py b/modules/api/models.py index 96cfe920..4cd20a92 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -209,11 +209,9 @@ class PreprocessResponse(BaseModel): fields = {} for key, metadata in opts.data_labels.items(): value = opts.data.get(key) - if key == 'sd_model_checkpoint': - value = None - optType = opts.typemap.get(type(metadata.default), type(value)) + optType = opts.typemap.get(type(metadata.default), type(metadata.default)) - if isinstance(optType, types.NoneType): + if metadata.default is None: pass elif metadata is not None: fields.update({key: (Optional[optType], Field(default=metadata.default, description=metadata.label))}) diff --git a/modules/cmd_args.py b/modules/cmd_args.py index ae78f469..e401f641 100644 --- a/modules/cmd_args.py +++ b/modules/cmd_args.py @@ -15,6 +15,7 @@ parser.add_argument("--update-check", action='store_true', help="launch.py argum parser.add_argument("--test-server", action='store_true', help="launch.py argument: configure server for testing") parser.add_argument("--skip-prepare-environment", action='store_true', help="launch.py argument: skip all environment preparation") parser.add_argument("--skip-install", action='store_true', help="launch.py argument: skip installation of packages") +parser.add_argument("--do-not-download-clip", action='store_true', help="do not download CLIP model even if it's not included in the checkpoint") parser.add_argument("--data-dir", type=str, default=os.path.dirname(os.path.dirname(os.path.realpath(__file__))), help="base path where all user data is stored") parser.add_argument("--config", type=str, default=sd_default_config, 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",) diff --git a/modules/sd_models.py b/modules/sd_models.py index 5813b550..fb31a793 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -494,7 +494,7 @@ def load_model(checkpoint_info=None, already_loaded_state_dict=None): sd_model = None try: - with sd_disable_initialization.DisableInitialization(disable_clip=clip_is_included_into_sd): + with sd_disable_initialization.DisableInitialization(disable_clip=clip_is_included_into_sd or shared.cmd_opts.do_not_download_clip): sd_model = instantiate_from_config(sd_config.model) except Exception: pass From 7f7db1700bda40ba3171a49b6a4ef38f868b7d0a Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 18:16:23 +0300 Subject: [PATCH 030/147] linter fix --- modules/api/models.py | 1 - 1 file changed, 1 deletion(-) diff --git a/modules/api/models.py b/modules/api/models.py index 4cd20a92..bf97b1a3 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -1,5 +1,4 @@ import inspect -import types from pydantic import BaseModel, Field, create_model from typing import Any, Optional From 136c8859a49a35cbffe269aafc0bbdfca0b3561d Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 18 Jul 2023 20:11:30 +0300 Subject: [PATCH 031/147] add backwards compatibility --lyco-dir-backcompat option, use that for LyCORIS directory instead of hardcoded value prevent running preload.py for disabled extensions --- CHANGELOG.md | 4 +--- extensions-builtin/Lora/networks.py | 4 ++-- extensions-builtin/Lora/preload.py | 1 + extensions-builtin/Lora/ui_extra_networks_lora.py | 4 ++-- launch.py | 1 + modules/script_loading.py | 5 +++-- modules/shared.py | 3 ++- 7 files changed, 12 insertions(+), 10 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 007010da..792529ec 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -58,9 +58,7 @@ * fix: check fill size none zero when resize (fixes #11425) * use submit and blur for quick settings textbox * save img2img batch with images.save_image() - * - - + * prevent running preload.py for disabled extensions ## 1.4.1 diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index 7b4c0312..af8188e3 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -11,7 +11,7 @@ import network_full import torch from typing import Union -from modules import shared, devices, sd_models, errors, scripts, sd_hijack, paths +from modules import shared, devices, sd_models, errors, scripts, sd_hijack module_types = [ network_lora.ModuleTypeLora(), @@ -399,7 +399,7 @@ def list_available_networks(): os.makedirs(shared.cmd_opts.lora_dir, exist_ok=True) candidates = list(shared.walk_files(shared.cmd_opts.lora_dir, allowed_extensions=[".pt", ".ckpt", ".safetensors"])) - candidates += list(shared.walk_files(os.path.join(paths.models_path, "LyCORIS"), allowed_extensions=[".pt", ".ckpt", ".safetensors"])) + candidates += list(shared.walk_files(shared.cmd_opts.lyco_dir_backcompat, allowed_extensions=[".pt", ".ckpt", ".safetensors"])) for filename in candidates: if os.path.isdir(filename): continue diff --git a/extensions-builtin/Lora/preload.py b/extensions-builtin/Lora/preload.py index 863dc5c0..50961be3 100644 --- a/extensions-builtin/Lora/preload.py +++ b/extensions-builtin/Lora/preload.py @@ -4,3 +4,4 @@ from modules import paths def preload(parser): parser.add_argument("--lora-dir", type=str, help="Path to directory with Lora networks.", default=os.path.join(paths.models_path, 'Lora')) + parser.add_argument("--lyco-dir-backcompat", type=str, help="Path to directory with LyCORIS networks (for backawards compatibility; can also use --lyco-dir).", default=os.path.join(paths.models_path, 'LyCORIS')) diff --git a/extensions-builtin/Lora/ui_extra_networks_lora.py b/extensions-builtin/Lora/ui_extra_networks_lora.py index 4b32098b..3629e5c0 100644 --- a/extensions-builtin/Lora/ui_extra_networks_lora.py +++ b/extensions-builtin/Lora/ui_extra_networks_lora.py @@ -3,7 +3,7 @@ import os import network import networks -from modules import shared, ui_extra_networks, paths +from modules import shared, ui_extra_networks from modules.ui_extra_networks import quote_js from ui_edit_user_metadata import LoraUserMetadataEditor @@ -72,7 +72,7 @@ class ExtraNetworksPageLora(ui_extra_networks.ExtraNetworksPage): yield item def allowed_directories_for_previews(self): - return [shared.cmd_opts.lora_dir, os.path.join(paths.models_path, "LyCORIS")] + return [shared.cmd_opts.lora_dir, shared.cmd_opts.lyco_dir_backcompat] def create_user_metadata_editor(self, ui, tabname): return LoraUserMetadataEditor(ui, tabname, self) diff --git a/launch.py b/launch.py index b103c8f3..1dbc4c6e 100644 --- a/launch.py +++ b/launch.py @@ -18,6 +18,7 @@ run_pip = launch_utils.run_pip check_run_python = launch_utils.check_run_python git_clone = launch_utils.git_clone git_pull_recursive = launch_utils.git_pull_recursive +list_extensions = launch_utils.list_extensions run_extension_installer = launch_utils.run_extension_installer prepare_environment = launch_utils.prepare_environment configure_for_tests = launch_utils.configure_for_tests diff --git a/modules/script_loading.py b/modules/script_loading.py index 306a1f35..0d55f193 100644 --- a/modules/script_loading.py +++ b/modules/script_loading.py @@ -12,11 +12,12 @@ def load_module(path): return module -def preload_extensions(extensions_dir, parser): +def preload_extensions(extensions_dir, parser, extension_list=None): if not os.path.isdir(extensions_dir): return - for dirname in sorted(os.listdir(extensions_dir)): + extensions = extension_list if extension_list is not None else os.listdir(extensions_dir) + for dirname in sorted(extensions): preload_script = os.path.join(extensions_dir, dirname, "preload.py") if not os.path.isfile(preload_script): continue diff --git a/modules/shared.py b/modules/shared.py index 6162938a..1ce7b49e 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -11,6 +11,7 @@ import gradio as gr import torch import tqdm +import launch import modules.interrogate import modules.memmon import modules.styles @@ -26,7 +27,7 @@ demo = None parser = cmd_args.parser -script_loading.preload_extensions(extensions_dir, parser) +script_loading.preload_extensions(extensions_dir, parser, extension_list=launch.list_extensions(launch.args.ui_settings_file)) script_loading.preload_extensions(extensions_builtin_dir, parser) if os.environ.get('IGNORE_CMD_ARGS_ERRORS', None) is None: From 2b42f73e3d198f7d21de3bba9253fdcd4cc0d76d Mon Sep 17 00:00:00 2001 From: kopyl Date: Tue, 18 Jul 2023 22:43:18 +0300 Subject: [PATCH 032/147] Make possible to install web ui without venv with --novenv flag When passing `--novenv` flag to webui.sh it can skip venv. Might be useful for installing in Docker since messing with venv in Docker might be a bit complicated. Example usage: `webui.sh --novenv` Hope this gets approved and pushed into future versions of Web UI --- webui.sh | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/webui.sh b/webui.sh index 5c8d977c..9ec14168 100755 --- a/webui.sh +++ b/webui.sh @@ -4,6 +4,11 @@ # change the variables in webui-user.sh instead # ################################################# +usevenv=1 +if [[ $@ == *"--novenv"* ]]; then + usevenv=0 +fi + # If run from macOS, load defaults from webui-macos-env.sh if [[ "$OSTYPE" == "darwin"* ]]; then if [[ -f webui-macos-env.sh ]] @@ -45,7 +50,7 @@ then fi # python3 venv without trailing slash (defaults to ${install_dir}/${clone_dir}/venv) -if [[ -z "${venv_dir}" ]] +if [[ -z "${venv_dir}" ]] && [[ $usevenv -eq 1 ]] then venv_dir="venv" fi @@ -158,7 +163,7 @@ do fi done -if ! "${python_cmd}" -c "import venv" &>/dev/null +if [[ $usevenv -eq 1 ]] && ! "${python_cmd}" -c "import venv" &>/dev/null then printf "\n%s\n" "${delimiter}" printf "\e[1m\e[31mERROR: python3-venv is not installed, aborting...\e[0m" @@ -178,7 +183,7 @@ else cd "${clone_dir}"/ || { printf "\e[1m\e[31mERROR: Can't cd to %s/%s/, aborting...\e[0m" "${install_dir}" "${clone_dir}"; exit 1; } fi -if [[ -z "${VIRTUAL_ENV}" ]]; +if [[ $usevenv -eq 1 ]] && [[ -z "${VIRTUAL_ENV}" ]]; then printf "\n%s\n" "${delimiter}" printf "Create and activate python venv" @@ -201,7 +206,7 @@ then fi else printf "\n%s\n" "${delimiter}" - printf "python venv already activate: ${VIRTUAL_ENV}" + printf "python venv already activate or run without venv: ${VIRTUAL_ENV}" printf "\n%s\n" "${delimiter}" fi From b010eea520caa90d2a31d98ec7c4a9d9d540c9ad Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 19 Jul 2023 00:41:00 +0300 Subject: [PATCH 033/147] fix incorrect multiplier for Loras --- extensions-builtin/Lora/extra_networks_lora.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/extensions-builtin/Lora/extra_networks_lora.py b/extensions-builtin/Lora/extra_networks_lora.py index 084c41d0..ba2945c6 100644 --- a/extensions-builtin/Lora/extra_networks_lora.py +++ b/extensions-builtin/Lora/extra_networks_lora.py @@ -25,7 +25,7 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork): te_multiplier = float(params.positional[1]) if len(params.positional) > 1 else 1.0 te_multiplier = float(params.named.get("te", te_multiplier)) - unet_multiplier = float(params.positional[2]) if len(params.positional) > 2 else 1.0 + unet_multiplier = float(params.positional[2]) if len(params.positional) > 2 else te_multiplier unet_multiplier = float(params.named.get("unet", unet_multiplier)) dyn_dim = int(params.positional[3]) if len(params.positional) > 3 else None From 0c4ca5f43e5e74aa1e7f65b3b6c20a676e158328 Mon Sep 17 00:00:00 2001 From: kopyl Date: Wed, 19 Jul 2023 01:47:39 +0300 Subject: [PATCH 034/147] Replace argument with env variable --- webui.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/webui.sh b/webui.sh index 9ec14168..5b39244b 100755 --- a/webui.sh +++ b/webui.sh @@ -5,7 +5,7 @@ ################################################# usevenv=1 -if [[ $@ == *"--novenv"* ]]; then +if [[ $venv_dir == "-" ]]; then usevenv=0 fi From 6094310704f4b3853bfa5d05d9c1ace58b2deee7 Mon Sep 17 00:00:00 2001 From: kopyl Date: Wed, 19 Jul 2023 01:48:21 +0300 Subject: [PATCH 035/147] improve var naming --- webui.sh | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/webui.sh b/webui.sh index 5b39244b..ed4cb4c8 100755 --- a/webui.sh +++ b/webui.sh @@ -4,9 +4,9 @@ # change the variables in webui-user.sh instead # ################################################# -usevenv=1 +use_venv=1 if [[ $venv_dir == "-" ]]; then - usevenv=0 + use_venv=0 fi # If run from macOS, load defaults from webui-macos-env.sh @@ -50,7 +50,7 @@ then fi # python3 venv without trailing slash (defaults to ${install_dir}/${clone_dir}/venv) -if [[ -z "${venv_dir}" ]] && [[ $usevenv -eq 1 ]] +if [[ -z "${venv_dir}" ]] && [[ $use_venv -eq 1 ]] then venv_dir="venv" fi @@ -163,7 +163,7 @@ do fi done -if [[ $usevenv -eq 1 ]] && ! "${python_cmd}" -c "import venv" &>/dev/null +if [[ $use_venv -eq 1 ]] && ! "${python_cmd}" -c "import venv" &>/dev/null then printf "\n%s\n" "${delimiter}" printf "\e[1m\e[31mERROR: python3-venv is not installed, aborting...\e[0m" @@ -183,7 +183,7 @@ else cd "${clone_dir}"/ || { printf "\e[1m\e[31mERROR: Can't cd to %s/%s/, aborting...\e[0m" "${install_dir}" "${clone_dir}"; exit 1; } fi -if [[ $usevenv -eq 1 ]] && [[ -z "${VIRTUAL_ENV}" ]]; +if [[ $use_venv -eq 1 ]] && [[ -z "${VIRTUAL_ENV}" ]]; then printf "\n%s\n" "${delimiter}" printf "Create and activate python venv" From c8b55f29e2838e67bd9e394f5dbca4350ccbb68f Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Wed, 19 Jul 2023 08:27:19 +0900 Subject: [PATCH 036/147] missing p save_image before-highres-fix --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/processing.py b/modules/processing.py index 6567b3cf..b89ca5c2 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1029,7 +1029,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): image = sd_samplers.sample_to_image(image, index, approximation=0) 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, suffix="-before-highres-fix") + 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: for i in range(samples.shape[0]): From cb7573489670cc7a042d24285e158b797c9558b2 Mon Sep 17 00:00:00 2001 From: yfzhou Date: Wed, 19 Jul 2023 17:53:28 +0800 Subject: [PATCH 037/147] =?UTF-8?q?=E3=80=90bug=E3=80=91reload=20altclip?= =?UTF-8?q?=20model=20error?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When using BertSeriesModelWithTransformation as the cond_stage_model, the undo_hijack should be performed using the FrozenXLMREmbedderWithCustomWords type; otherwise, it will result in a failed model reload. --- modules/sd_hijack.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/sd_hijack.py b/modules/sd_hijack.py index 3b6f95ce..928233ab 100644 --- a/modules/sd_hijack.py +++ b/modules/sd_hijack.py @@ -203,7 +203,7 @@ class StableDiffusionModelHijack: ldm.modules.diffusionmodules.openaimodel.UNetModel.forward = sd_unet.UNetModel_forward def undo_hijack(self, m): - if type(m.cond_stage_model) == xlmr.BertSeriesModelWithTransformation: + if type(m.cond_stage_model) == sd_hijack_xlmr.FrozenXLMREmbedderWithCustomWords: m.cond_stage_model = m.cond_stage_model.wrapped elif type(m.cond_stage_model) == sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords: From 05ccb4d0e3ff38503a2863889e21290e77590ee2 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Wed, 19 Jul 2023 15:49:31 +0300 Subject: [PATCH 038/147] bugfix: model name was added together with directory name to infotext and to [model_name] filename pattern --- CHANGELOG.md | 1 + modules/images.py | 2 +- modules/processing.py | 2 +- 3 files changed, 3 insertions(+), 2 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 792529ec..a561252c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -59,6 +59,7 @@ * use submit and blur for quick settings textbox * save img2img batch with images.save_image() * prevent running preload.py for disabled extensions + * fix: previously, model name was added together with directory name to infotext and to [model_name] filename pattern; directory name is now not included ## 1.4.1 diff --git a/modules/images.py b/modules/images.py index fb5d2e75..38aa933d 100644 --- a/modules/images.py +++ b/modules/images.py @@ -363,7 +363,7 @@ class FilenameGenerator: 'styles': lambda self: self.p and sanitize_filename_part(", ".join([style for style in self.p.styles if not style == "None"]) or "None", replace_spaces=False), 'sampler': lambda self: self.p and sanitize_filename_part(self.p.sampler_name, replace_spaces=False), 'model_hash': lambda self: getattr(self.p, "sd_model_hash", shared.sd_model.sd_model_hash), - 'model_name': lambda self: sanitize_filename_part(shared.sd_model.sd_checkpoint_info.model_name, replace_spaces=False), + 'model_name': lambda self: sanitize_filename_part(shared.sd_model.sd_checkpoint_info.name_for_extra, replace_spaces=False), 'date': lambda self: datetime.datetime.now().strftime('%Y-%m-%d'), 'datetime': lambda self, *args: self.datetime(*args), # accepts formats: [datetime], [datetime], [datetime