(self, n_samples, rand=None, return_idx=False)

(self, n_samples, rand=None)

(self, x: torch.Tensor, sigma: float)

(self, x, sigma, scale=None)

(self, x, sigma, step_index, scale=None)

(self, x, sigma)

(self, input_ids, scores)

(self, input_ids, scores)

(self)

(self, args, **kwargs)

( self, in_channels, hidden_size, patch_size, bias=True, text_

(self, height, width, compressed_num_frames, hidden_size, text_length=0)

( self, height, width, compressed_num_frames, hidden_size, tex

( self, height, width, compressed_num_frames, hidden_size, hid

( self, hidden_size, time_embed_dim, patch_size, out_channels,

(self, num_layers, in_features, hidden_features, bias=False)

( self, transformer_args, num_frames, time_compressed_rate, latent_wid

(self, path, process_fn, seed, *, shuffle_buffer=1000)

( self, path, process_fn, seed, *, meta_names=[], nshards=sys.maxsize, shuffle_buffer=1000, include_di

( self, warm_up_steps, lr_min, lr_max, lr_start, max_decay_ste

(self, warm_up_steps, f_min, f_max, f_start, cycle_lengths, verbosity_interval=0)

(self, seeds, weights)

( self, in_channels, n_heads, d_head, depth=1, dropout=0.0,

(self, model, decay=0.9999, use_num_upates=True)

(self, dim, dim_out=None, mult=4, glu=False, dropout=0.0)

(self, dim, heads=4, dim_head=32)

(self, in_channels)

( self, query_dim, context_dim=None, heads=8, dim_head=64, dro

( self, query_dim, context_dim=None, heads=8, dim_head=64, dro

( self, query_dim, context_dim=None, heads=8, dim_head=64, dro

(self, *, dim, dim_head=64, heads=8)

( self, dim=1920, depth=4, dim_head=64, heads=16, num_queries=

(self, query_dim, context_dim=None, heads=8, dim_head=64, dropout=0.0, **kwargs)

( self, dim, n_heads, d_head, dropout=0.0, context_dim=None,

( self, dim, n_heads, d_head, dropout=0.0, context_dim=None,

( self, in_channels, n_heads, d_head, depth=1, dropout=0.0,

( self, sigma_sampler_config, type="l2", offset_noise_level=0.0, batch

(self, sigma_data=0.5)

(self)

( self, spacial_dim: int, embed_dim: int, num_heads_channels: int, out

(self, channels, use_conv, dims=2, out_channels=None, padding=1, third_up=False)

(self, channels, out_channels=None, ks=5)

(self, channels, use_conv, dims=2, out_channels=None, padding=1, third_down=False)

( self, channels, emb_channels, dropout, out_channels=None, us

( self, channels, num_heads=1, num_head_channels=-1, use_checkpoint=Fa

(self, n_heads)

(self, n_heads)

(self, dim)

( self, in_channels, model_channels, out_channels, num_res_blocks,

( self, image_size, in_channels, model_channels, out_channels,

(self, sigma_data: float = 0.5)

(self, scale_factor)

( self, linear_start=0.00085, linear_end=0.0120, num_timesteps=1000, )

( self, linear_start=0.00085, linear_end=0.0120, num_timesteps=1000, s

( self, discretization_config: Union[Dict, ListConfig, OmegaConf], num_steps: Union[in

(self, s_noise=0.1, *args, **kwargs)

(self, eta=1.0, s_noise=1.0, *args, **kwargs)

( self, order=4, *args, **kwargs, )

(self, edit_ratio=0.5, *args, **kwargs)

(self, fixed_frames=0, sdedit=False, **kwargs)

( self, alpha: float, merge_strategy: str = "learned_with_images", rearrange_p

(self, p_mean=-1.2, p_std=1.2)

(self, discretization_config, num_idx, do_append_zero=False, flip=True, uniform_sampling=False)

(self, discretization_config, total_num_idx, partial_num_idx, do_append_zero=False, flip=True)

( self, weighting_config, scaling_config, num_idx, discretization_conf

( self, in_features, out_features, rank=4, kernel_size=(1, 1), stride=(1, 1), padding=0, network_alpha

(self, *args, lora_layer: Optional[LoRAConv2dLayer] = None, scale: float = 1.0, **kwargs)

(self, *args, lora_layer: Optional[LoRALinearLayer] = None, scale: float = 1.0, **kwargs)

(self, diffusion_model, compile_model: bool = False, dtype: torch.dtype = torch.float32)

(self, in_channels, with_conv)

(self, in_channels, with_conv)

( self, *, in_channels, out_channels=None, conv_shortcut=False,

(self, in_channels)

(self, in_channels)

(self, in_channels)

( self, *, ch, out_ch, ch_mult=(1, 2, 4, 8), num_res_blocks,

( self, *, ch, out_ch, ch_mult=(1, 2, 4, 8), num_res_blocks,

(self, scale, exp, num_steps, dyn_thresh_config=None)

(self, fn: Module)

(self, fn: Module)

(self, dim, *, dim_out=None, dim_hidden_min=16, init_bias=-10)

(self, fn: Module)

(self, dim, channel_first=False, images=False, bias=False)

(self, dim, *, dim_cond, channel_first=False, images=False, bias=False)

( self, *, dim, dim_cond: Optional[int] = None, causal=False,

(self, *, dim, dim_cond: Optional[int] = None, dim_head=8, heads=8, dropout=0.0)

(self, dim, *, dim_cond: Optional[int] = None, mult=4, images=False)

(self, input_channels, filters, downsample=True, antialiased_downsample=True)

( self, *, dim, image_size, channels=3, max_dim=512, a

( self, dim, *, spatial_kernel, time_kernel, causal=True, dim_out=None, demod=True, eps=1e-8, pad_mode

(self, dim, dim_out=None, kernel_size=3, antialias=False)

(self, dim, dim_out=None, kernel_size=3, antialias=False)

(self, dim, dim_out=None)

(self, dim, dim_out=None)

( self, chan_in, chan_out, kernel_size: Union[int, Tuple[int, int, int]], pad_mode="constant", **kwarg

( self, dim, kernel_size: Union[int, Tuple[int, int, int]], *, dim_cond, pad_mode: str = "constant", d

(self, chan_in, chan_out, kernel_size: Union[int, Tuple[int, int, int]], *, time_stride, **kwargs)

( self, *, image_size, layers: Tuple[Union[str, Tuple[str, int]], ...] = ("res

(self)

( self, out_channels, *args, dropout=0.0, video_kernel_size=3,