(block)

(self, filename)

(self, filename_text, idx=None)

(self, dim, drop_path=0., layer_scale_init_value=1e-6)

(self, idx=None)

(embedding, checkpoint, embedding_name, filename, remove_cached_checksum=True, use_negative=False, embedding_n

(seed)

(self)

(self, ma_model, current_model)

(self, x)

(cfg_text:str)

(self, img)

(module, prefix='')

(model, state_dict, prefix='', ignore_missing="relative_position_index")

(self, embedding, model)

(self, old, new)

(model_name, learn_rate, batch_size, data_root, template_file, steps, save_model_every, create_image_every, lo

(log_directory, filename, step, epoch_len, values)

(self, image)

(self, i)

(self, i)

(self, in_chans=3, num_classes=1000, depths=[3, 3, 9, 3], dims=[96, 192, 384, 768], drop_pat

(self, normalized_shape, eps=1e-6, data_format="channels_last")

(self, filename=None, latent=None, filename_text=None, timg=None, att_mask=None)

(self, width, height)

(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template

(self, data_root, width, height, device=None)

(self, ckpt)

(self, vec, name, step=None)

(self, embeddings_dir)

(self, beta)

(self)

(self)

(self, m)

(module, x, output)

(self)

(a)

(pretrained=False, in_22k=False, **kwargs)

(pretrained=False, in_22k=False, **kwargs)

(pretrained=False,in_22k=False, **kwargs)

(pretrained=False,in_22k=False, **kwargs)

(pretrained=False,in_22k=False, pretrained_cfg=None, **kwargs)

(pretrained=False, in_22k=False, **kwargs)

(name, initialization_text, initialization_text_neg, nvpt, overwrite_old, nvpt_neg, seed)

(name, num_vectors_per_token, overwrite_old, init_text='*')

(self, tokens, offset)

(self, x)

(self, x)

(self, x)

(self, img)

(self)

(current_step)

()

(params)

(x_start, cond, t, noise=None, scale=(1.0,1.0), att_mask=None, dy_cfg_f='ln')

(*args)

(id_task, data_root, training_width, training_height)

(path, filename)

()

(*args)

(id_task, embedding_name, seed, learn_rate, batch_size, data_root, log_directory, training_width, training_hei