From e21f01f64504bc651da6e85216474bbd35ee010d Mon Sep 17 00:00:00 2001 From: Rae Fu Date: Thu, 13 Oct 2022 23:00:38 -0600 Subject: add checkpoint cache option to UI for faster model switching switching time reduced from ~1500ms to ~280ms --- modules/sd_models.py | 54 +++++++++++++++++++++++++++++++--------------------- modules/shared.py | 1 + 2 files changed, 33 insertions(+), 22 deletions(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 0a55b4c3..f3660d8d 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -1,4 +1,4 @@ -import glob +import collections import os.path import sys from collections import namedtuple @@ -15,6 +15,7 @@ model_path = os.path.abspath(os.path.join(models_path, model_dir)) CheckpointInfo = namedtuple("CheckpointInfo", ['filename', 'title', 'hash', 'model_name', 'config']) checkpoints_list = {} +checkpoints_loaded = collections.OrderedDict() try: # this silences the annoying "Some weights of the model checkpoint were not used when initializing..." message at start. @@ -132,38 +133,46 @@ def load_model_weights(model, checkpoint_info): checkpoint_file = checkpoint_info.filename sd_model_hash = checkpoint_info.hash - print(f"Loading weights [{sd_model_hash}] from {checkpoint_file}") + if checkpoint_info not in checkpoints_loaded: + print(f"Loading weights [{sd_model_hash}] from {checkpoint_file}") - pl_sd = torch.load(checkpoint_file, map_location="cpu") - if "global_step" in pl_sd: - print(f"Global Step: {pl_sd['global_step']}") + pl_sd = torch.load(checkpoint_file, map_location="cpu") + if "global_step" in pl_sd: + print(f"Global Step: {pl_sd['global_step']}") - sd = get_state_dict_from_checkpoint(pl_sd) + sd = get_state_dict_from_checkpoint(pl_sd) + model.load_state_dict(sd, strict=False) - model.load_state_dict(sd, strict=False) + if shared.cmd_opts.opt_channelslast: + model.to(memory_format=torch.channels_last) - if shared.cmd_opts.opt_channelslast: - model.to(memory_format=torch.channels_last) + if not shared.cmd_opts.no_half: + model.half() - if not shared.cmd_opts.no_half: - model.half() + devices.dtype = torch.float32 if shared.cmd_opts.no_half else torch.float16 + devices.dtype_vae = torch.float32 if shared.cmd_opts.no_half or shared.cmd_opts.no_half_vae else torch.float16 - devices.dtype = torch.float32 if shared.cmd_opts.no_half else torch.float16 - devices.dtype_vae = torch.float32 if shared.cmd_opts.no_half or shared.cmd_opts.no_half_vae else torch.float16 + vae_file = os.path.splitext(checkpoint_file)[0] + ".vae.pt" - vae_file = os.path.splitext(checkpoint_file)[0] + ".vae.pt" + if not os.path.exists(vae_file) and shared.cmd_opts.vae_path is not None: + vae_file = shared.cmd_opts.vae_path - if not os.path.exists(vae_file) and shared.cmd_opts.vae_path is not None: - vae_file = shared.cmd_opts.vae_path + if os.path.exists(vae_file): + print(f"Loading VAE weights from: {vae_file}") + vae_ckpt = torch.load(vae_file, map_location="cpu") + vae_dict = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss"} - if os.path.exists(vae_file): - print(f"Loading VAE weights from: {vae_file}") - vae_ckpt = torch.load(vae_file, map_location="cpu") - vae_dict = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss"} + model.first_stage_model.load_state_dict(vae_dict) - model.first_stage_model.load_state_dict(vae_dict) + model.first_stage_model.to(devices.dtype_vae) - model.first_stage_model.to(devices.dtype_vae) + checkpoints_loaded[checkpoint_info] = model.state_dict().copy() + while len(checkpoints_loaded) > shared.opts.sd_checkpoint_cache: + checkpoints_loaded.popitem(last=False) # LRU + else: + print(f"Loading weights [{sd_model_hash}] from cache") + checkpoints_loaded.move_to_end(checkpoint_info) + model.load_state_dict(checkpoints_loaded[checkpoint_info]) model.sd_model_hash = sd_model_hash model.sd_model_checkpoint = checkpoint_file @@ -202,6 +211,7 @@ def reload_model_weights(sd_model, info=None): return if sd_model.sd_checkpoint_info.config != checkpoint_info.config: + checkpoints_loaded.clear() shared.sd_model = load_model() return shared.sd_model diff --git a/modules/shared.py b/modules/shared.py index 5901e605..b2090da1 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -238,6 +238,7 @@ 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": modules.sd_models.checkpoint_tiles()}, refresh=sd_models.list_models), + "sd_checkpoint_cache": OptionInfo(0, "Checkpoints to cache in RAM", gr.Slider, {"minimum": 0, "maximum": 10, "step": 1}), "sd_hypernetwork": OptionInfo("None", "Hypernetwork", gr.Dropdown, lambda: {"choices": ["None"] + [x for x in hypernetworks.keys()]}, refresh=reload_hypernetworks), "sd_hypernetwork_strength": OptionInfo(1.0, "Hypernetwork strength", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.001}), "img2img_color_correction": OptionInfo(False, "Apply color correction to img2img results to match original colors."), -- cgit v1.2.3 From 4d19f3b7d461fe0f63e7ccff936909b0ce0c6126 Mon Sep 17 00:00:00 2001 From: Melan Date: Fri, 14 Oct 2022 22:45:26 +0200 Subject: Raise an assertion error if no training images have been found. --- modules/textual_inversion/dataset.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py index 67e90afe..12e2f43b 100644 --- a/modules/textual_inversion/dataset.py +++ b/modules/textual_inversion/dataset.py @@ -81,7 +81,8 @@ class PersonalizedBase(Dataset): entry.cond = cond_model([entry.cond_text]).to(devices.cpu) self.dataset.append(entry) - + + assert len(self.dataset) > 1, "No images have been found in the dataset." self.length = len(self.dataset) * repeats self.initial_indexes = np.arange(self.length) % len(self.dataset) -- cgit v1.2.3 From c7a86f7fe9c0b8967a87e8d709f507d2f44400d8 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 09:24:59 +0300 Subject: add option to use batch size for training --- modules/hypernetworks/hypernetwork.py | 33 +++++++++++++++++++------- modules/textual_inversion/dataset.py | 31 ++++++++++++++---------- modules/textual_inversion/textual_inversion.py | 17 +++++++------ modules/ui.py | 3 +++ 4 files changed, 54 insertions(+), 30 deletions(-) (limited to 'modules') diff --git a/modules/hypernetworks/hypernetwork.py b/modules/hypernetworks/hypernetwork.py index 59c7ac6e..a2b3bc0a 100644 --- a/modules/hypernetworks/hypernetwork.py +++ b/modules/hypernetworks/hypernetwork.py @@ -182,7 +182,21 @@ def attention_CrossAttention_forward(self, x, context=None, mask=None): return self.to_out(out) -def train_hypernetwork(hypernetwork_name, learn_rate, data_root, log_directory, steps, create_image_every, save_hypernetwork_every, template_file, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height): +def stack_conds(conds): + if len(conds) == 1: + return torch.stack(conds) + + # same as in reconstruct_multicond_batch + token_count = max([x.shape[0] for x in conds]) + for i in range(len(conds)): + if conds[i].shape[0] != token_count: + last_vector = conds[i][-1:] + last_vector_repeated = last_vector.repeat([token_count - conds[i].shape[0], 1]) + conds[i] = torch.vstack([conds[i], last_vector_repeated]) + + return torch.stack(conds) + +def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log_directory, steps, create_image_every, save_hypernetwork_every, template_file, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height): assert hypernetwork_name, 'hypernetwork not selected' path = shared.hypernetworks.get(hypernetwork_name, None) @@ -211,7 +225,7 @@ def train_hypernetwork(hypernetwork_name, learn_rate, data_root, log_directory, shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..." with torch.autocast("cuda"): - ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=512, height=512, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=hypernetwork_name, model=shared.sd_model, device=devices.device, template_file=template_file, include_cond=True) + ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=512, height=512, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=hypernetwork_name, model=shared.sd_model, device=devices.device, template_file=template_file, include_cond=True, batch_size=batch_size) if unload: shared.sd_model.cond_stage_model.to(devices.cpu) @@ -235,7 +249,7 @@ def train_hypernetwork(hypernetwork_name, learn_rate, data_root, log_directory, optimizer = torch.optim.AdamW(weights, lr=scheduler.learn_rate) pbar = tqdm.tqdm(enumerate(ds), total=steps - ititial_step) - for i, entry in pbar: + for i, entries in pbar: hypernetwork.step = i + ititial_step scheduler.apply(optimizer, hypernetwork.step) @@ -246,11 +260,12 @@ def train_hypernetwork(hypernetwork_name, learn_rate, data_root, log_directory, break with torch.autocast("cuda"): - cond = entry.cond.to(devices.device) - x = entry.latent.to(devices.device) - loss = shared.sd_model(x.unsqueeze(0), cond)[0] + c = stack_conds([entry.cond for entry in entries]).to(devices.device) +# c = torch.vstack([entry.cond for entry in entries]).to(devices.device) + x = torch.stack([entry.latent for entry in entries]).to(devices.device) + loss = shared.sd_model(x, c)[0] del x - del cond + del c losses[hypernetwork.step % losses.shape[0]] = loss.item() @@ -292,7 +307,7 @@ def train_hypernetwork(hypernetwork_name, learn_rate, data_root, log_directory, p.width = preview_width p.height = preview_height else: - p.prompt = entry.cond_text + p.prompt = entries[0].cond_text p.steps = 20 preview_text = p.prompt @@ -315,7 +330,7 @@ def train_hypernetwork(hypernetwork_name, learn_rate, data_root, log_directory,

Loss: {losses.mean():.7f}
Step: {hypernetwork.step}
-Last prompt: {html.escape(entry.cond_text)}
+Last prompt: {html.escape(entries[0].cond_text)}
Last saved embedding: {html.escape(last_saved_file)}
Last saved image: {html.escape(last_saved_image)}

diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py index 67e90afe..bd99c0cb 100644 --- a/modules/textual_inversion/dataset.py +++ b/modules/textual_inversion/dataset.py @@ -24,11 +24,12 @@ class DatasetEntry: class PersonalizedBase(Dataset): - def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None, include_cond=False): - re_word = re.compile(shared.opts.dataset_filename_word_regex) if len(shared.opts.dataset_filename_word_regex)>0 else None + def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None, include_cond=False, batch_size=1): + re_word = re.compile(shared.opts.dataset_filename_word_regex) if len(shared.opts.dataset_filename_word_regex) > 0 else None self.placeholder_token = placeholder_token + self.batch_size = batch_size self.width = width self.height = height self.flip = transforms.RandomHorizontalFlip(p=flip_p) @@ -78,13 +79,13 @@ class PersonalizedBase(Dataset): if include_cond: entry.cond_text = self.create_text(filename_text) - entry.cond = cond_model([entry.cond_text]).to(devices.cpu) + entry.cond = cond_model([entry.cond_text]).to(devices.cpu).squeeze(0) self.dataset.append(entry) - self.length = len(self.dataset) * repeats + self.length = len(self.dataset) * repeats // batch_size - self.initial_indexes = np.arange(self.length) % len(self.dataset) + self.initial_indexes = np.arange(len(self.dataset)) self.indexes = None self.shuffle() @@ -101,13 +102,19 @@ class PersonalizedBase(Dataset): return self.length def __getitem__(self, i): - if i % len(self.dataset) == 0: - self.shuffle() + res = [] - index = self.indexes[i % len(self.indexes)] - entry = self.dataset[index] + for j in range(self.batch_size): + position = i * self.batch_size + j + if position % len(self.indexes) == 0: + self.shuffle() - if entry.cond is None: - entry.cond_text = self.create_text(entry.filename_text) + index = self.indexes[position % len(self.indexes)] + entry = self.dataset[index] - return entry + if entry.cond is None: + entry.cond_text = self.create_text(entry.filename_text) + + res.append(entry) + + return res diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index da0d77a0..e754747e 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -199,7 +199,7 @@ def write_loss(log_directory, filename, step, epoch_len, values): }) -def train_embedding(embedding_name, learn_rate, data_root, log_directory, training_width, training_height, steps, create_image_every, save_embedding_every, template_file, save_image_with_stored_embedding, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height): +def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_directory, training_width, training_height, steps, create_image_every, save_embedding_every, template_file, save_image_with_stored_embedding, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height): assert embedding_name, 'embedding not selected' shared.state.textinfo = "Initializing textual inversion training..." @@ -231,7 +231,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..." with torch.autocast("cuda"): - ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file) + ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file, batch_size=batch_size) hijack = sd_hijack.model_hijack @@ -251,7 +251,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini optimizer = torch.optim.AdamW([embedding.vec], lr=scheduler.learn_rate) pbar = tqdm.tqdm(enumerate(ds), total=steps-ititial_step) - for i, entry in pbar: + for i, entries in pbar: embedding.step = i + ititial_step scheduler.apply(optimizer, embedding.step) @@ -262,10 +262,9 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini break with torch.autocast("cuda"): - c = cond_model([entry.cond_text]) - - x = entry.latent.to(devices.device) - loss = shared.sd_model(x.unsqueeze(0), c)[0] + c = cond_model([entry.cond_text for entry in entries]) + x = torch.stack([entry.latent for entry in entries]).to(devices.device) + loss = shared.sd_model(x, c)[0] del x losses[embedding.step % losses.shape[0]] = loss.item() @@ -307,7 +306,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini p.width = preview_width p.height = preview_height else: - p.prompt = entry.cond_text + p.prompt = entries[0].cond_text p.steps = 20 p.width = training_width p.height = training_height @@ -348,7 +347,7 @@ def train_embedding(embedding_name, learn_rate, data_root, log_directory, traini

Loss: {losses.mean():.7f}
Step: {embedding.step}
-Last prompt: {html.escape(entry.cond_text)}
+Last prompt: {html.escape(entries[0].cond_text)}
Last saved embedding: {html.escape(last_saved_file)}
Last saved image: {html.escape(last_saved_image)}

diff --git a/modules/ui.py b/modules/ui.py index 1bc919c7..45550ea8 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -1166,6 +1166,7 @@ def create_ui(wrap_gradio_gpu_call): train_embedding_name = gr.Dropdown(label='Embedding', choices=sorted(sd_hijack.model_hijack.embedding_db.word_embeddings.keys())) train_hypernetwork_name = gr.Dropdown(label='Hypernetwork', choices=[x for x in shared.hypernetworks.keys()]) learn_rate = gr.Textbox(label='Learning rate', placeholder="Learning rate", value="0.005") + batch_size = gr.Number(label='Batch size', value=1, precision=0) dataset_directory = gr.Textbox(label='Dataset directory', placeholder="Path to directory with input images") log_directory = gr.Textbox(label='Log directory', placeholder="Path to directory where to write outputs", value="textual_inversion") template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt")) @@ -1244,6 +1245,7 @@ def create_ui(wrap_gradio_gpu_call): inputs=[ train_embedding_name, learn_rate, + batch_size, dataset_directory, log_directory, training_width, @@ -1268,6 +1270,7 @@ def create_ui(wrap_gradio_gpu_call): inputs=[ train_hypernetwork_name, learn_rate, + batch_size, dataset_directory, log_directory, steps, -- cgit v1.2.3 From db27b987a97fc8b7894a9dd34bd7641536f9c424 Mon Sep 17 00:00:00 2001 From: aoirusann Date: Sat, 15 Oct 2022 11:48:13 +0800 Subject: Add hint for `ctrl/alt enter` And duplicate implementations are removed --- modules/ui.py | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index 45550ea8..baf4c397 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -433,7 +433,10 @@ def create_toprow(is_img2img): with gr.Row(): with gr.Column(scale=80): with gr.Row(): - prompt = gr.Textbox(label="Prompt", elem_id=f"{id_part}_prompt", show_label=False, placeholder="Prompt", lines=2) + prompt = gr.Textbox(label="Prompt", elem_id=f"{id_part}_prompt", show_label=False, lines=2, + placeholder="Prompt (press Ctrl+Enter or Alt+Enter to generate)" + ) + with gr.Column(scale=1, elem_id="roll_col"): roll = gr.Button(value=art_symbol, elem_id="roll", visible=len(shared.artist_db.artists) > 0) paste = gr.Button(value=paste_symbol, elem_id="paste") @@ -446,7 +449,10 @@ def create_toprow(is_img2img): with gr.Row(): with gr.Column(scale=8): with gr.Row(): - negative_prompt = gr.Textbox(label="Negative prompt", elem_id="negative_prompt", show_label=False, placeholder="Negative prompt", lines=2) + negative_prompt = gr.Textbox(label="Negative prompt", elem_id=f"{id_part}_neg_prompt", show_label=False, lines=2, + placeholder="Negative prompt (press Ctrl+Enter or Alt+Enter to generate)" + ) + with gr.Column(scale=1, elem_id="roll_col"): sh = gr.Button(elem_id="sh", visible=True) -- cgit v1.2.3 From cd28465bf87d911965790513c37e6881e4231523 Mon Sep 17 00:00:00 2001 From: ddPn08 Date: Sat, 15 Oct 2022 10:56:02 +0900 Subject: do not force relative paths in image history --- modules/images_history.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/images_history.py b/modules/images_history.py index f5ef44fe..09c749fe 100644 --- a/modules/images_history.py +++ b/modules/images_history.py @@ -104,7 +104,7 @@ def show_images_history(gr, opts, tabname, run_pnginfo, switch_dict): elif tabname == "extras": dir_name = opts.outdir_extras_samples d = dir_name.split("/") - dir_name = d[0] + dir_name = "/" if dir_name.startswith("/") else d[0] for p in d[1:]: dir_name = os.path.join(dir_name, p) with gr.Row(): -- cgit v1.2.3 From 0da6c1809996f0f696d4047faf4b9c9939e26daa Mon Sep 17 00:00:00 2001 From: ddPn08 Date: Sat, 15 Oct 2022 11:22:05 +0900 Subject: use "outdir_samples" if specified --- modules/images_history.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/images_history.py b/modules/images_history.py index 09c749fe..9260df8a 100644 --- a/modules/images_history.py +++ b/modules/images_history.py @@ -97,7 +97,9 @@ def delete_image(delete_num, tabname, dir_name, name, page_index, filenames, ima def show_images_history(gr, opts, tabname, run_pnginfo, switch_dict): - if tabname == "txt2img": + if opts.outdir_samples != "": + dir_name = opts.outdir_samples + elif tabname == "txt2img": dir_name = opts.outdir_txt2img_samples elif tabname == "img2img": dir_name = opts.outdir_img2img_samples -- cgit v1.2.3 From a13af34b902bebc5df9509228380206a01f1245b Mon Sep 17 00:00:00 2001 From: githublsx Date: Thu, 13 Oct 2022 20:05:07 -0700 Subject: Set to -1 when seed input is none --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index a75b9f84..7e2a416d 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -140,7 +140,7 @@ class Processed: self.sampler_noise_scheduler_override = p.sampler_noise_scheduler_override self.prompt = self.prompt if type(self.prompt) != list else self.prompt[0] self.negative_prompt = self.negative_prompt if type(self.negative_prompt) != list else self.negative_prompt[0] - self.seed = int(self.seed if type(self.seed) != list else self.seed[0]) + self.seed = int(self.seed if type(self.seed) != list else self.seed[0]) if self.seed is not None else -1 self.subseed = int(self.subseed if type(self.subseed) != list else self.subseed[0]) if self.subseed is not None else -1 self.all_prompts = all_prompts or [self.prompt] -- cgit v1.2.3 From c24df4b486a48c60f48276f7760a9acb4a13e22d Mon Sep 17 00:00:00 2001 From: CookieHCl Date: Sat, 15 Oct 2022 03:26:36 +0900 Subject: Disable compiling deepbooru model This is only necessary when you have to train, and compiling model produces warning. --- modules/deepbooru.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/deepbooru.py b/modules/deepbooru.py index f34f3788..4ad334a1 100644 --- a/modules/deepbooru.py +++ b/modules/deepbooru.py @@ -102,7 +102,7 @@ def get_deepbooru_tags_model(): tags = dd.project.load_tags_from_project(model_path) model = dd.project.load_model_from_project( - model_path, compile_model=True + model_path, compile_model=False ) return model, tags -- cgit v1.2.3 From e8729dd0511f8410db967d9ef192645cfef1be8a Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 12:54:23 +0300 Subject: re-apply height hacks to work with new gradio --- modules/ui.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index baf4c397..9c7a67dd 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -750,10 +750,10 @@ def create_ui(wrap_gradio_gpu_call): with gr.Tabs(elem_id="mode_img2img") as tabs_img2img_mode: with gr.TabItem('img2img', id='img2img'): - init_img = gr.Image(label="Image for img2img", elem_id="img2img_image", show_label=False, source="upload", interactive=True, type="pil", tool=cmd_opts.gradio_img2img_tool) + init_img = gr.Image(label="Image for img2img", elem_id="img2img_image", show_label=False, source="upload", interactive=True, type="pil", tool=cmd_opts.gradio_img2img_tool).style(height=480) with gr.TabItem('Inpaint', id='inpaint'): - init_img_with_mask = gr.Image(label="Image for inpainting with mask", show_label=False, elem_id="img2maskimg", source="upload", interactive=True, type="pil", tool="sketch", image_mode="RGBA") + init_img_with_mask = gr.Image(label="Image for inpainting with mask", show_label=False, elem_id="img2maskimg", source="upload", interactive=True, type="pil", tool="sketch", image_mode="RGBA").style(height=480) init_img_inpaint = gr.Image(label="Image for img2img", show_label=False, source="upload", interactive=True, type="pil", visible=False, elem_id="img_inpaint_base") init_mask_inpaint = gr.Image(label="Mask", source="upload", interactive=True, type="pil", visible=False, elem_id="img_inpaint_mask") -- cgit v1.2.3 From 5967d07d1aa4e2fef031a57b1612b1ab04a3cd78 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 13:11:28 +0300 Subject: fix new gradio failing to preserve image params --- modules/ui.py | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index 9c7a67dd..de5ab929 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -7,6 +7,7 @@ import mimetypes import os import random import sys +import tempfile import time import traceback import platform @@ -176,6 +177,23 @@ def save_files(js_data, images, do_make_zip, index): return gr.File.update(value=fullfns, visible=True), '', '', plaintext_to_html(f"Saved: {filenames[0]}") +def save_pil_to_file(pil_image, dir=None): + use_metadata = False + metadata = PngImagePlugin.PngInfo() + for key, value in pil_image.info.items(): + if isinstance(key, str) and isinstance(value, str): + metadata.add_text(key, value) + use_metadata = True + + file_obj = tempfile.NamedTemporaryFile(delete=False, suffix=".png", dir=dir) + pil_image.save(file_obj, pnginfo=(metadata if use_metadata else None)) + return file_obj + + +# override save to file function so that it also writes PNG info +gr.processing_utils.save_pil_to_file = save_pil_to_file + + def wrap_gradio_call(func, extra_outputs=None): def f(*args, extra_outputs_array=extra_outputs, **kwargs): run_memmon = opts.memmon_poll_rate > 0 and not shared.mem_mon.disabled -- cgit v1.2.3 From f7ca63937ac83d32483285c3af09afaa356d6276 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 13:23:12 +0300 Subject: bring back scale latent option in settings --- modules/processing.py | 8 ++++---- modules/shared.py | 3 ++- 2 files changed, 6 insertions(+), 5 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 7e2a416d..b9a1660e 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -557,11 +557,11 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): samples = samples[:, :, self.truncate_y//2:samples.shape[2]-self.truncate_y//2, self.truncate_x//2:samples.shape[3]-self.truncate_x//2] - decoded_samples = decode_first_stage(self.sd_model, samples) + if opts.use_scale_latent_for_hires_fix: + samples = torch.nn.functional.interpolate(samples, size=(self.height // opt_f, self.width // opt_f), mode="bilinear") - if opts.upscaler_for_img2img is None or opts.upscaler_for_img2img == "None": - decoded_samples = torch.nn.functional.interpolate(decoded_samples, size=(self.height, self.width), mode="bilinear") else: + decoded_samples = decode_first_stage(self.sd_model, samples) lowres_samples = torch.clamp((decoded_samples + 1.0) / 2.0, min=0.0, max=1.0) batch_images = [] @@ -578,7 +578,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): decoded_samples = decoded_samples.to(shared.device) decoded_samples = 2. * decoded_samples - 1. - samples = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(decoded_samples)) + samples = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(decoded_samples)) shared.state.nextjob() diff --git a/modules/shared.py b/modules/shared.py index aa69bedf..b4141e67 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -218,6 +218,7 @@ options_templates.update(options_section(('upscaling', "Upscaling"), { "SWIN_tile_overlap": OptionInfo(8, "Tile overlap, in pixels for SwinIR. Low values = visible seam.", gr.Slider, {"minimum": 0, "maximum": 48, "step": 1}), "ldsr_steps": OptionInfo(100, "LDSR processing steps. Lower = faster", gr.Slider, {"minimum": 1, "maximum": 200, "step": 1}), "upscaler_for_img2img": OptionInfo(None, "Upscaler for img2img", gr.Dropdown, lambda: {"choices": [x.name for x in sd_upscalers]}), + "use_scale_latent_for_hires_fix": OptionInfo(False, "Upscale latent space iamge when doing hires. fix"), })) options_templates.update(options_section(('face-restoration', "Face restoration"), { @@ -256,7 +257,6 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), { "filter_nsfw": OptionInfo(False, "Filter NSFW content"), 'CLIP_stop_at_last_layers': OptionInfo(1, "Stop At last layers of CLIP model", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}), "random_artist_categories": OptionInfo([], "Allowed categories for random artists selection when using the Roll button", gr.CheckboxGroup, {"choices": artist_db.categories()}), - 'quicksettings': OptionInfo("sd_model_checkpoint", "Quicksettings list"), })) options_templates.update(options_section(('interrogate', "Interrogate Options"), { @@ -284,6 +284,7 @@ options_templates.update(options_section(('ui', "User interface"), { "js_modal_lightbox": OptionInfo(True, "Enable full page image viewer"), "js_modal_lightbox_initially_zoomed": OptionInfo(True, "Show images zoomed in by default in full page image viewer"), "show_progress_in_title": OptionInfo(True, "Show generation progress in window title."), + 'quicksettings': OptionInfo("sd_model_checkpoint", "Quicksettings list"), })) options_templates.update(options_section(('sampler-params', "Sampler parameters"), { -- cgit v1.2.3 From d3463bc59a44d62c2de8b357184c49876d84f654 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 14:22:30 +0300 Subject: change styling for top right corner UI made save style button not die when you cancel --- modules/ui.py | 57 ++++++++++++++++++++++++++++----------------------------- 1 file changed, 28 insertions(+), 29 deletions(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index de5ab929..cab8ab11 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -81,6 +81,8 @@ art_symbol = '\U0001f3a8' # 🎨 paste_symbol = '\u2199\ufe0f' # ↙ folder_symbol = '\U0001f4c2' # 📂 refresh_symbol = '\U0001f504' # 🔄 +save_style_symbol = '\U0001f4be' # 💾 +apply_style_symbol = '\U0001f4cb' # 📋 def plaintext_to_html(text): @@ -322,7 +324,7 @@ def visit(x, func, path=""): def add_style(name: str, prompt: str, negative_prompt: str): if name is None: - return [gr_show(), gr_show()] + return [gr_show() for x in range(4)] style = modules.styles.PromptStyle(name, prompt, negative_prompt) shared.prompt_styles.styles[style.name] = style @@ -447,7 +449,7 @@ def create_toprow(is_img2img): id_part = "img2img" if is_img2img else "txt2img" with gr.Row(elem_id="toprow"): - with gr.Column(scale=4): + with gr.Column(scale=6): with gr.Row(): with gr.Column(scale=80): with gr.Row(): @@ -455,27 +457,30 @@ def create_toprow(is_img2img): placeholder="Prompt (press Ctrl+Enter or Alt+Enter to generate)" ) - with gr.Column(scale=1, elem_id="roll_col"): - roll = gr.Button(value=art_symbol, elem_id="roll", visible=len(shared.artist_db.artists) > 0) - paste = gr.Button(value=paste_symbol, elem_id="paste") - token_counter = gr.HTML(value="", elem_id=f"{id_part}_token_counter") - token_button = gr.Button(visible=False, elem_id=f"{id_part}_token_button") - - with gr.Column(scale=10, elem_id="style_pos_col"): - prompt_style = gr.Dropdown(label="Style 1", elem_id=f"{id_part}_style_index", choices=[k for k, v in shared.prompt_styles.styles.items()], value=next(iter(shared.prompt_styles.styles.keys())), visible=len(shared.prompt_styles.styles) > 1) - with gr.Row(): - with gr.Column(scale=8): + with gr.Column(scale=80): with gr.Row(): negative_prompt = gr.Textbox(label="Negative prompt", elem_id=f"{id_part}_neg_prompt", show_label=False, lines=2, placeholder="Negative prompt (press Ctrl+Enter or Alt+Enter to generate)" ) - with gr.Column(scale=1, elem_id="roll_col"): - sh = gr.Button(elem_id="sh", visible=True) + with gr.Column(scale=1, elem_id="roll_col"): + roll = gr.Button(value=art_symbol, elem_id="roll", visible=len(shared.artist_db.artists) > 0) + paste = gr.Button(value=paste_symbol, elem_id="paste") + save_style = gr.Button(value=save_style_symbol, elem_id="style_create") + prompt_style_apply = gr.Button(value=apply_style_symbol, elem_id="style_apply") - with gr.Column(scale=1, elem_id="style_neg_col"): - prompt_style2 = gr.Dropdown(label="Style 2", elem_id=f"{id_part}_style2_index", choices=[k for k, v in shared.prompt_styles.styles.items()], value=next(iter(shared.prompt_styles.styles.keys())), visible=len(shared.prompt_styles.styles) > 1) + token_counter = gr.HTML(value="", elem_id=f"{id_part}_token_counter") + token_button = gr.Button(visible=False, elem_id=f"{id_part}_token_button") + + button_interrogate = None + button_deepbooru = None + if is_img2img: + with gr.Column(scale=1, elem_id="interrogate_col"): + button_interrogate = gr.Button('Interrogate\nCLIP', elem_id="interrogate") + + if cmd_opts.deepdanbooru: + button_deepbooru = gr.Button('Interrogate\nDeepBooru', elem_id="deepbooru") with gr.Column(scale=1): with gr.Row(): @@ -495,20 +500,14 @@ def create_toprow(is_img2img): outputs=[], ) - with gr.Row(scale=1): - if is_img2img: - interrogate = gr.Button('Interrogate\nCLIP', elem_id="interrogate") - if cmd_opts.deepdanbooru: - deepbooru = gr.Button('Interrogate\nDeepBooru', elem_id="deepbooru") - else: - deepbooru = None - else: - interrogate = None - deepbooru = None - prompt_style_apply = gr.Button('Apply style', elem_id="style_apply") - save_style = gr.Button('Create style', elem_id="style_create") + with gr.Row(): + with gr.Column(scale=1, elem_id="style_pos_col"): + prompt_style = gr.Dropdown(label="Style 1", elem_id=f"{id_part}_style_index", choices=[k for k, v in shared.prompt_styles.styles.items()], value=next(iter(shared.prompt_styles.styles.keys()))) + + with gr.Column(scale=1, elem_id="style_neg_col"): + prompt_style2 = gr.Dropdown(label="Style 2", elem_id=f"{id_part}_style2_index", choices=[k for k, v in shared.prompt_styles.styles.items()], value=next(iter(shared.prompt_styles.styles.keys()))) - return prompt, roll, prompt_style, negative_prompt, prompt_style2, submit, interrogate, deepbooru, prompt_style_apply, save_style, paste, token_counter, token_button + return prompt, roll, prompt_style, negative_prompt, prompt_style2, submit, button_interrogate, button_deepbooru, prompt_style_apply, save_style, paste, token_counter, token_button def setup_progressbar(progressbar, preview, id_part, textinfo=None): -- cgit v1.2.3 From 20a1f68c752f8e37492ea00911c97bfc572a6e67 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 15:44:46 +0300 Subject: fix gadio issue with sending files between tabs --- modules/ui.py | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index cab8ab11..c9b53247 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -91,6 +91,14 @@ def plaintext_to_html(text): def image_from_url_text(filedata): + if type(filedata) == dict and filedata["is_file"]: + filename = filedata["name"] + tempdir = os.path.normpath(tempfile.gettempdir()) + normfn = os.path.normpath(filename) + assert normfn.startswith(tempdir), 'trying to open image file not in temporary directory' + + return Image.open(filename) + if type(filedata) == list: if len(filedata) == 0: return None -- cgit v1.2.3 From 97f0727489ddd3d7ca264c54ed0f63b6847502e2 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 15:47:02 +0300 Subject: add First pass size always regardless of whether it was auto chosen or specified --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index b9a1660e..941ae089 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -528,7 +528,6 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): firstphase_height_truncated = int(scale * self.height) else: - self.extra_generation_params["First pass size"] = f"{self.firstphase_width}x{self.firstphase_height}" width_ratio = self.width / self.firstphase_width height_ratio = self.height / self.firstphase_height @@ -540,6 +539,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): firstphase_width_truncated = self.firstphase_height * self.width / self.height firstphase_height_truncated = self.firstphase_height + self.extra_generation_params["First pass size"] = f"{self.firstphase_width}x{self.firstphase_height}" self.truncate_x = int(self.firstphase_width - firstphase_width_truncated) // opt_f self.truncate_y = int(self.firstphase_height - firstphase_height_truncated) // opt_f -- cgit v1.2.3 From eef3bc649069d6caaef1274f132de28e528bfa7d Mon Sep 17 00:00:00 2001 From: NO_ob <15161159+NO-ob@users.noreply.github.com> Date: Sat, 15 Oct 2022 11:43:30 +0100 Subject: typo --- modules/shared.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/shared.py b/modules/shared.py index b4141e67..fa30bbb0 100644 --- a/modules/shared.py +++ b/modules/shared.py @@ -218,7 +218,7 @@ options_templates.update(options_section(('upscaling', "Upscaling"), { "SWIN_tile_overlap": OptionInfo(8, "Tile overlap, in pixels for SwinIR. Low values = visible seam.", gr.Slider, {"minimum": 0, "maximum": 48, "step": 1}), "ldsr_steps": OptionInfo(100, "LDSR processing steps. Lower = faster", gr.Slider, {"minimum": 1, "maximum": 200, "step": 1}), "upscaler_for_img2img": OptionInfo(None, "Upscaler for img2img", gr.Dropdown, lambda: {"choices": [x.name for x in sd_upscalers]}), - "use_scale_latent_for_hires_fix": OptionInfo(False, "Upscale latent space iamge when doing hires. fix"), + "use_scale_latent_for_hires_fix": OptionInfo(False, "Upscale latent space image when doing hires. fix"), })) options_templates.update(options_section(('face-restoration', "Face restoration"), { -- cgit v1.2.3 From 5fd638f14d75a71a37157ded5d33c716ab9eb8ca Mon Sep 17 00:00:00 2001 From: ruocaled Date: Sat, 15 Oct 2022 02:00:46 -0700 Subject: fix download section layout --- modules/ui.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index c9b53247..3206113e 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -619,7 +619,7 @@ def create_ui(wrap_gradio_gpu_call): txt2img_preview = gr.Image(elem_id='txt2img_preview', visible=False) txt2img_gallery = gr.Gallery(label='Output', show_label=False, elem_id='txt2img_gallery').style(grid=4) - with gr.Group(): + with gr.Column(): with gr.Row(): save = gr.Button('Save') send_to_img2img = gr.Button('Send to img2img') @@ -834,7 +834,7 @@ def create_ui(wrap_gradio_gpu_call): img2img_preview = gr.Image(elem_id='img2img_preview', visible=False) img2img_gallery = gr.Gallery(label='Output', show_label=False, elem_id='img2img_gallery').style(grid=4) - with gr.Group(): + with gr.Column(): with gr.Row(): save = gr.Button('Save') img2img_send_to_img2img = gr.Button('Send to img2img') -- cgit v1.2.3 From 703e6d9e4e161d36b9328eefb5200e1c44fb4afd Mon Sep 17 00:00:00 2001 From: AngelBottomless <35677394+aria1th@users.noreply.github.com> Date: Sat, 15 Oct 2022 21:47:08 +0900 Subject: check NaN for hypernetwork tuning --- modules/hypernetworks/hypernetwork.py | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/hypernetworks/hypernetwork.py b/modules/hypernetworks/hypernetwork.py index a2b3bc0a..4905710e 100644 --- a/modules/hypernetworks/hypernetwork.py +++ b/modules/hypernetworks/hypernetwork.py @@ -272,15 +272,17 @@ def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log optimizer.zero_grad() loss.backward() optimizer.step() - - pbar.set_description(f"loss: {losses.mean():.7f}") + mean_loss = losses.mean() + if torch.isnan(mean_loss): + raise RuntimeError("Loss diverged.") + pbar.set_description(f"loss: {mean_loss:.7f}") if hypernetwork.step > 0 and hypernetwork_dir is not None and hypernetwork.step % save_hypernetwork_every == 0: last_saved_file = os.path.join(hypernetwork_dir, f'{hypernetwork_name}-{hypernetwork.step}.pt') hypernetwork.save(last_saved_file) textual_inversion.write_loss(log_directory, "hypernetwork_loss.csv", hypernetwork.step, len(ds), { - "loss": f"{losses.mean():.7f}", + "loss": f"{mean_loss:.7f}", "learn_rate": scheduler.learn_rate }) @@ -328,7 +330,7 @@ def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log shared.state.textinfo = f"""

-Loss: {losses.mean():.7f}
+Loss: {mean_loss:.7f}
Step: {hypernetwork.step}
Last prompt: {html.escape(entries[0].cond_text)}
Last saved embedding: {html.escape(last_saved_file)}
-- cgit v1.2.3 From 9e846083b702a498fdb60accd72f075fa26701d9 Mon Sep 17 00:00:00 2001 From: DepFA <35278260+dfaker@users.noreply.github.com> Date: Fri, 14 Oct 2022 14:50:25 +0100 Subject: add vector size to embed text --- modules/textual_inversion/textual_inversion.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index e754747e..6f549d62 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -327,10 +327,16 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc info.add_text("sd-ti-embedding", embedding_to_b64(data)) title = "<{}>".format(data.get('name', '???')) + + try: + vectorSize = list(data['string_to_param'].values())[0].shape[0] + except Exception as e: + vectorSize = '?' + checkpoint = sd_models.select_checkpoint() footer_left = checkpoint.model_name footer_mid = '[{}]'.format(checkpoint.hash) - footer_right = '{}'.format(embedding.step) + footer_right = 'v{} {}s'.format(vectorSize, embedding.step) captioned_image = caption_image_overlay(image, title, footer_left, footer_mid, footer_right) captioned_image = insert_image_data_embed(captioned_image, data) -- cgit v1.2.3 From 939f16529a72fe48c2ce3ef31bdaba785925a33c Mon Sep 17 00:00:00 2001 From: DepFA <35278260+dfaker@users.noreply.github.com> Date: Fri, 14 Oct 2022 14:55:05 +0100 Subject: only save 1 image per embedding --- modules/textual_inversion/textual_inversion.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index 6f549d62..1d697c90 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -242,6 +242,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc last_saved_file = "" last_saved_image = "" + embedding_yet_to_be_embedded = False ititial_step = embedding.step or 0 if ititial_step > steps: @@ -281,6 +282,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc if embedding.step > 0 and embedding_dir is not None and embedding.step % save_embedding_every == 0: last_saved_file = os.path.join(embedding_dir, f'{embedding_name}-{embedding.step}.pt') embedding.save(last_saved_file) + embedding_yet_to_be_embedded = True write_loss(log_directory, "textual_inversion_loss.csv", embedding.step, len(ds), { "loss": f"{losses.mean():.7f}", @@ -318,7 +320,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc shared.state.current_image = image - if save_image_with_stored_embedding and os.path.exists(last_saved_file): + if save_image_with_stored_embedding and os.path.exists(last_saved_file) and embedding_yet_to_be_embedded: last_saved_image_chunks = os.path.join(images_embeds_dir, f'{embedding_name}-{embedding.step}.png') @@ -342,6 +344,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc captioned_image = insert_image_data_embed(captioned_image, data) captioned_image.save(last_saved_image_chunks, "PNG", pnginfo=info) + embedding_yet_to_be_embedded = False image.save(last_saved_image) -- cgit v1.2.3 From 9a1dcd78edbf9caf68b9e6286d7b5ca81500e243 Mon Sep 17 00:00:00 2001 From: DepFA <35278260+dfaker@users.noreply.github.com> Date: Fri, 14 Oct 2022 18:14:02 +0100 Subject: add webp for embed load --- modules/textual_inversion/textual_inversion.py | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'modules') diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index 1d697c90..c07bffc3 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -96,6 +96,10 @@ class EmbeddingDatabase: else: data = extract_image_data_embed(embed_image) name = data.get('name', name) + elif filename.upper().endswith('.WEBP'): + embed_image = Image.open(path) + data = extract_image_data_embed(embed_image) + name = data.get('name', name) else: data = torch.load(path, map_location="cpu") -- cgit v1.2.3 From ddf6899df0cf87d4da77cb2ce223061f4a5edf18 Mon Sep 17 00:00:00 2001 From: DepFA <35278260+dfaker@users.noreply.github.com> Date: Fri, 14 Oct 2022 18:23:20 +0100 Subject: generalise to popular lossless formats --- modules/textual_inversion/textual_inversion.py | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'modules') diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index c07bffc3..b99df3b1 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -88,18 +88,14 @@ class EmbeddingDatabase: data = [] - if filename.upper().endswith('.PNG'): + if os.path.splitext(filename.upper())[-1] in ['.PNG', '.WEBP', '.JXL', '.AVIF']: embed_image = Image.open(path) - if 'sd-ti-embedding' in embed_image.text: + if hasattr(embed_image, 'text') and 'sd-ti-embedding' in embed_image.text: data = embedding_from_b64(embed_image.text['sd-ti-embedding']) name = data.get('name', name) else: data = extract_image_data_embed(embed_image) name = data.get('name', name) - elif filename.upper().endswith('.WEBP'): - embed_image = Image.open(path) - data = extract_image_data_embed(embed_image) - name = data.get('name', name) else: data = torch.load(path, map_location="cpu") -- cgit v1.2.3 From b6e3b96dab94a00f51725f9cc977eebc6b4072ab Mon Sep 17 00:00:00 2001 From: DepFA <35278260+dfaker@users.noreply.github.com> Date: Sat, 15 Oct 2022 15:17:21 +0100 Subject: Change vector size footer label --- modules/textual_inversion/textual_inversion.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index b99df3b1..2ed345b1 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -338,7 +338,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc checkpoint = sd_models.select_checkpoint() footer_left = checkpoint.model_name footer_mid = '[{}]'.format(checkpoint.hash) - footer_right = 'v{} {}s'.format(vectorSize, embedding.step) + footer_right = '{}v {}s'.format(vectorSize, embedding.step) captioned_image = caption_image_overlay(image, title, footer_left, footer_mid, footer_right) captioned_image = insert_image_data_embed(captioned_image, data) -- cgit v1.2.3 From 74a9ee70020ffa2746c82300c533de3f7e523f22 Mon Sep 17 00:00:00 2001 From: AUTOMATIC <16777216c@gmail.com> Date: Sat, 15 Oct 2022 17:25:35 +0300 Subject: fix saving images compatibility with gradio update --- modules/ui.py | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index 3206113e..b867d40f 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -154,10 +154,7 @@ def save_files(js_data, images, do_make_zip, index): writer.writerow(["prompt", "seed", "width", "height", "sampler", "cfgs", "steps", "filename", "negative_prompt"]) for image_index, filedata in enumerate(images, start_index): - if filedata.startswith("data:image/png;base64,"): - filedata = filedata[len("data:image/png;base64,"):] - - image = Image.open(io.BytesIO(base64.decodebytes(filedata.encode('utf-8')))) + image = image_from_url_text(filedata) is_grid = image_index < p.index_of_first_image i = 0 if is_grid else (image_index - p.index_of_first_image) -- cgit v1.2.3 From 529afbf4d70165a0dfd19eb9c2ec22416b794a1d Mon Sep 17 00:00:00 2001 From: C43H66N12O12S2 <36072735+C43H66N12O12S2@users.noreply.github.com> Date: Sat, 15 Oct 2022 19:19:54 +0300 Subject: Update sd_hijack.py --- modules/sd_hijack.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_hijack.py b/modules/sd_hijack.py index c81722a0..984b35c4 100644 --- a/modules/sd_hijack.py +++ b/modules/sd_hijack.py @@ -24,7 +24,7 @@ def apply_optimizations(): ldm.modules.diffusionmodules.model.nonlinearity = silu - if cmd_opts.force_enable_xformers or (cmd_opts.xformers and shared.xformers_available and torch.version.cuda and (6, 0) <= torch.cuda.get_device_capability(shared.device) <= (8, 6)): + if cmd_opts.force_enable_xformers or (cmd_opts.xformers and shared.xformers_available and torch.version.cuda and (6, 0) <= torch.cuda.get_device_capability(shared.device) <= (9, 0)): print("Applying xformers cross attention optimization.") ldm.modules.attention.CrossAttention.forward = sd_hijack_optimizations.xformers_attention_forward ldm.modules.diffusionmodules.model.AttnBlock.forward = sd_hijack_optimizations.xformers_attnblock_forward -- cgit v1.2.3