(self, *args, **kwargs)

( self, in_channels: int, out_channels: int, factor_t, factor_s=1,

(state_dict, with_shape=True)

(self, x, scale)

(self, video, torch_dtype=None, device=None, pattern="B C T H W", min_value=-1, max_value=1)

(self, in_dim, out_dim, has_pos_emb=False)

(model, module_class)

(self)

(self, model_names=[])

(self, *args, **kwargs)

(self, dim, eps=1e-6)

(module)

(self, x)

(self, z, scale)

(self, model_name)

( torch_dtype = torch.bfloat16, device = "cuda", vae_model_path = None, text_e

(self, item)

(text)

(self, file_path, model_type=None, device=None, torch_dtype=None)

(file_path, torch_dtype=None, device="cpu")

(dim: int, end: int = 1024, theta: float = 10000.0)

(self, csv_path, dataset_base_path)

(self, file_path)

(self, num_inference_steps=100, denoising_strength=1.0, training=False, shift=None, dynamic_shift_len=None, ex

()

(self, model_output, timestep, sample, to_final=False, **kwargs)

(self)

(self, length, left_bound, right_bound, border_width)

(self, data, is_bound, border_width)

(self, height, width, num_frames=None)

(model)

(self, prompt, positive=True, device="cuda")

(self, state_dict, remove_prefix=None)

(self, tokenizer_path=None)

(q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, num_heads: int, compatibility_mode=False)

(self, x: torch.Tensor, first_chunk=False)

(x)

(self, shape, seed=None, rand_device="cpu", rand_torch_dtype=torch.float32, device=None, torch_dtype=None)

(self, image)

(self, model: torch.nn.Module, state_dict_lora, alpha=1.0)

(x: torch.Tensor, shift: torch.Tensor, scale: torch.Tensor)

(self, x: torch.Tensor, control_camera_latents_input: Optional[torch.Tensor] = None)

(self, inputs)

(x, freqs, num_heads)

(self, accelerator, model, file_name)

(dim, position)

(self, x: torch.Tensor, grid_size: torch.Tensor)

(text)

( self, vace_layers=(0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28), vace_in_d

(self, text)

(self, rel_pos)

( self, model, target_modules, lora_rank, upcast_d

(self, original_samples, noise, timestep)

( self, image_seq_len, base_seq_len: int = 256, max_seq_len: int = 8192,

(text, keep_punctuation_exact_string=None)

(state_dict, with_shape=True)

(state_dict, with_shape=True)

(self, text_encoder: WanTextEncoder = None)

( self, x, vace_context, context, t_mod, freqs, use_gradient_checkpointing: bool = False,

(self, data)

(self, model_names)

(self, state_dict)

(self, lora_state_dict)

(state_dict, with_shape=True)

(device = torch.device("meta"), include_buffers :bool = False)

(self, inputs)

(self, inputs)

( dataset, model, model_logger, learning_rate = 1e-5, weight_decay = 1e-2, num_workers

( self, module, lora_config = None, alpha = 1 )

(state_dict, model_names, model_classes, torch_dtype, device)

(file_path, torch_dtype=None, device="cpu")

(file_path, torch_dtype=None, device="cpu")

(self, inputs)

(self, inputs)

(self, accelerator, model, epoch_id)

(self, accelerator, model, save_steps=None)

(self, accelerator, model, save_steps=None)

()

(fn)

(x, patch_size)

(dim: int, end: int = 1024, theta: float = 10000.0)

(self, image, torch_dtype=None, device=None, pattern="B C H W", min_value=-1, max_value=1)

(self, file_path)

(self, prompt, positive=True)

(self, inputs)

(self, mu, log_var)

(frames, save_path, fps, quality=9, ffmpeg_params=None)

(self, height, width)

(self, inputs)

(self, inputs)

(self, hidden_state, device)

(self, video, device)

( self, pipe, lora_base_model, # "vace" lora_target_modules, # "q,k,

(self, hidden_states, device, tile_size, tile_stride)

(self, video, device, tile_size, tile_stride)

(self)

(self)

(self, **inputs)

(self, sample, noise, timestep)

(self, timestep)