mirror/stable-diffusion-webui
2
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-01-06 14:35:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add checkbox for variable training dims

Browse Source
This commit is contained in:
dan 2023-01-08 01:34:52 +08:00
parent 448b9cedab
commit 669fb18d52
4 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
modules/hypernetworks/hypernetwork.py
View File

@ -403,7 +403,7 @@ def create_hypernetwork(name, enable_sizes, overwrite_old, layer_structure=None,
shared.reload_hypernetworks() shared.reload_hypernetworks()
def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, 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 train_hypernetwork(hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, varsize, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, 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):
# images allows training previews to have infotext. Importing it at the top causes a circular import problem. # images allows training previews to have infotext. Importing it at the top causes a circular import problem.
from modules import images from modules import images

4
modules/textual_inversion/dataset.py
View File

@ -29,7 +29,7 @@ class DatasetEntry:
class PersonalizedBase(Dataset): class PersonalizedBase(Dataset):
def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, cond_model=None, device=None, template_file=None, include_cond=False, batch_size=1, gradient_step=1, shuffle_tags=False, tag_drop_out=0, latent_sampling_method='once'): def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, cond_model=None, device=None, template_file=None, include_cond=False, batch_size=1, gradient_step=1, shuffle_tags=False, tag_drop_out=0, latent_sampling_method='once', varsize=False):
re_word = re.compile(shared.opts.dataset_filename_word_regex) if len(shared.opts.dataset_filename_word_regex) > 0 else None 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.placeholder_token = placeholder_token
@ -59,7 +59,7 @@ class PersonalizedBase(Dataset):
raise Exception("interrupted") raise Exception("interrupted")
try: try:
image = Image.open(path).convert('RGB') image = Image.open(path).convert('RGB')
if width < 2000: if not varsize:
image = image.resize((width, height), PIL.Image.BICUBIC) image = image.resize((width, height), PIL.Image.BICUBIC)
else: else:
assert batch_size == 1, 'variable img size must have batch size 1' assert batch_size == 1, 'variable img size must have batch size 1'

4
modules/textual_inversion/textual_inversion.py
View File

@ -255,7 +255,7 @@ def validate_train_inputs(model_name, learn_rate, batch_size, gradient_step, dat
if save_model_every or create_image_every: if save_model_every or create_image_every:
assert log_directory, "Log directory is empty" assert log_directory, "Log directory is empty"
def train_embedding(embedding_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, 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, gradient_step, data_root, log_directory, training_width, training_height, varsize, steps, clip_grad_mode, clip_grad_value, shuffle_tags, tag_drop_out, latent_sampling_method, 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):
save_embedding_every = save_embedding_every or 0 save_embedding_every = save_embedding_every or 0
create_image_every = create_image_every or 0 create_image_every = create_image_every or 0
validate_train_inputs(embedding_name, learn_rate, batch_size, gradient_step, data_root, template_file, steps, save_embedding_every, create_image_every, log_directory, name="embedding") validate_train_inputs(embedding_name, learn_rate, batch_size, gradient_step, data_root, template_file, steps, save_embedding_every, create_image_every, log_directory, name="embedding")
@ -309,7 +309,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, gradient_step, data_
pin_memory = shared.opts.pin_memory pin_memory = shared.opts.pin_memory
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, cond_model=shared.sd_model.cond_stage_model, device=devices.device, template_file=template_file, batch_size=batch_size, gradient_step=gradient_step, shuffle_tags=shuffle_tags, tag_drop_out=tag_drop_out, latent_sampling_method=latent_sampling_method) 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, cond_model=shared.sd_model.cond_stage_model, device=devices.device, template_file=template_file, batch_size=batch_size, gradient_step=gradient_step, shuffle_tags=shuffle_tags, tag_drop_out=tag_drop_out, latent_sampling_method=latent_sampling_method, varsize=varsize)
if shared.opts.save_training_settings_to_txt: if shared.opts.save_training_settings_to_txt:
save_settings_to_file(log_directory, {**dict(model_name=checkpoint.model_name, model_hash=checkpoint.hash, num_of_dataset_images=len(ds), num_vectors_per_token=len(embedding.vec)), **locals()}) save_settings_to_file(log_directory, {**dict(model_name=checkpoint.model_name, model_hash=checkpoint.hash, num_of_dataset_images=len(ds), num_vectors_per_token=len(embedding.vec)), **locals()})

3
modules/ui.py
View File

@ -1343,6 +1343,7 @@ def create_ui():
template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"), elem_id="train_template_file") template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"), elem_id="train_template_file")
training_width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="train_training_width") training_width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="train_training_width")
training_height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="train_training_height") training_height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="train_training_height")
varsize = gr.Checkbox(label="Ignore dimension settings and do not resize images", value=False, elem_id="train_varsize")
steps = gr.Number(label='Max steps', value=100000, precision=0, elem_id="train_steps") steps = gr.Number(label='Max steps', value=100000, precision=0, elem_id="train_steps")
with FormRow(): with FormRow():
@ -1449,6 +1450,7 @@ def create_ui():
log_directory, log_directory,
training_width, training_width,
training_height, training_height,
varsize,
steps, steps,
clip_grad_mode, clip_grad_mode,
clip_grad_value, clip_grad_value,
@ -1480,6 +1482,7 @@ def create_ui():
log_directory, log_directory,
training_width, training_width,
training_height, training_height,
varsize,
steps, steps,
clip_grad_mode, clip_grad_mode,
clip_grad_value, clip_grad_value,