( self, unet: fl.Module, lda: LatentDiffusionAutoencoder, clip_text_encoder: f

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, device: Device | str | None = None, dtype: DType | None = None)

( self, channels: int, num_attention_layers: int = 1, num_attention_heads: int

(self, in_channels: int, device: Device | str | None = None, dtype: DType | None = None)

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, in_channels: int, device: Device | str | None = None, dtype: DType | None = None)

(self, target: SDXLUNet, scale: float = 1.0, kernel_size: int = 9, sigma: float = 1.0)

( self, target: SDXLUNet, clip_image_encoder: CLIPImageEncoderH | None = None,

( self, target: SDXLUNet, name: str, condition_encoder: ConditionEncoderXL | N

( self, unet: SDXLUNet | None = None, lda: SDXLAutoencoder | None = None, clip

( self, target: CLIPTextEncoderG, projection: fl.Linear | None = None, )

( self, num_inference_steps: int, num_train_timesteps: int = 1_000, initial_di

( self, num_inference_steps: int, num_train_timesteps: int = 1_000, initial_di

( self, num_inference_steps: int, num_train_timesteps: int = 1_000, initial_di

( self, num_inference_steps: int, num_train_timesteps: int = 1_000, initial_di

( self, num_inference_steps: int, num_train_timesteps: int = 1_000, initial_di

( self, in_channels: int = 3, # RGB out_channels: int = 1, # Grayscale n_res

( self, context_key: str = "timestep_embedding", device: Device | str | None = None,

( self, in_channels: int, out_channels: int, num_groups: int = 32, eps

( self, channels: int, device: Device | str | None = None, dtype: DType | None

( self, in_channels: int, device: Device | str | None = None, dtype: DType | N

( self, device: Device | str | None = None, dtype: DType | None = None, )

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, n: int)

(self, n: int)

(self, target: SD1UNet, scale: float = 1.0, kernel_size: int = 9, sigma: float = 1.0)

( self, target: SD1UNet, clip_image_encoder: CLIPImageEncoderH | None = None,

(self, device: Device | str | None = None, dtype: DType | None = None)

( self, target: SD1UNet, name: str, scale: float = 1.0, weights: dict[str, Tensor] | None = None )

( self, target: SD1UNet, name: str, condition_encoder: ConditionEncoder | None

( self, unet: SD1UNet | None = None, lda: SD1Autoencoder | None = None, clip_t

(self, num_output_mask: int = 3)

( self, embedding_dim: int = 256, num_layers: int = 3, num_mask_tokens: int =

( self, embedding_dim: int = 256, dense_embedding_side_dim: int = 64, device:

( self, embedding_dim: int = 256, num_mask_tokens: int = 4, device: Device | s

( self, embedding_dim: int, num_mask_tokens: int, num_layers: int = 3,

( self, embedding_dim: int, num_layers: int, num_mask_tokens: int, dev

( self, in_channels: int, out_channels: int, patch_size: int = 16, use

( self, embedding_dim: int, image_embedding_size: tuple[int, int], device: Dev

( self, embedding_dim: int, num_heads: int, spatial_size: tuple[int, int],

( self, embedding_dim: int = 768, spatial_size: tuple[int, int] = (64, 64), nu

( self, embedding_dim: int, feedforward_dim: int, device: Device | str | None

(self)

(self)

( self, embedding_dim: int, num_heads: int, feedforward_dim: int, imag

( self, embedding_dim: int, num_layers: int, num_heads: int, global_at

(self, device: Device | str | None = None, dtype: DType | None = None)

( self, embedding_dim: int, cross_embedding_dim: int | None = None, num_heads:

( self, embedding_dim: int, inner_dim: int | None = None, num_heads: int = 1,

( self, embedding_dim: int, num_heads: int = 8, device: Device | str | None = None, dtype: DType | Non

( self, embedding_dim: int, num_heads: int = 8, device: Device | str | None = None, dtype: DType | Non

( self, embedding_dim: int, num_heads: int = 8, feed_forward_dim: int = 2048,

( self, num_positional_features: int = 64, scale: float = 1, device: Device |

(self, embedding_dim: int, device: Device | str | None = None, dtype: DType | None = None)

( self, embedding_dim: int = 256, scale: float = 1, device: Device | str | None = None, dtype: DType |

( self, image_encoder: SAMViTH | None = None, point_encoder: PointEncoder | None = Non

( self, vocabulary_path: str | Path = Path(__file__).resolve().parent / "bpe_simple_vocab_16e6

( self, in_channels: int, out_channels: int, patch_size: int = 16, use

( self, embedding_dim: int = 768, feedforward_dim: int = 3072, num_attention_h

( self, image_size: int = 224, embedding_dim: int = 768, patch_size: int = 32,

( self, image_size: int = 224, embedding_dim: int = 768, output_dim: int = 512

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, device: Device | str | None = None, dtype: DType | None = None)

( self, target: TokenEncoder, )

(self, target: CLIPTextEncoder)

( self, max_sequence_length: int, embedding_dim: int, device: Device | str | N

( self, embedding_dim: int, feedforward_dim: int, num_attention_heads: int = 1

( self, embedding_dim: int = 768, max_sequence_length: int = 77, vocabulary_si

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, device: Device | str | None = None, dtype: DType | None = None)

(self, device: Device | str | None = None, dtype: DType | None = None)

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, embedding_dim: int, device: torch.device | str | None = None, dtype: t

( self, sequence_length: int, embedding_dim: int, device: torch.device | str |

( self, embedding_dim: int, init_value: float = 1.0, dtype: torch.dtype | None

( self, embedding_dim: int, feedforward_dim: int, activation: Activation = fl.

( self, in_channels: int, out_channels: int, patch_size: int, device:

( self, embedding_dim: int, num_heads: int, norm_eps: float, mlp_ratio

( self, num_registers: int, embedding_dim: int, device: torch.device | str | N

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, device: torch.device | str | None = None, dtype: torch.dtype | None = None,

( self, num_pos_feats=64, temperature=10000, normalize=False, scale=None )

( self, embed_dim, nhead=8, dim_mlp=2048, dropout=0.0, activation="gelu" )

(self, in_ch, out_ch)

(self, codebook_size, emb_dim, beta)

( self, codebook_size, emb_dim, num_hiddens, straight_through=False,

(self, in_channels)

(self, in_channels)

(self, in_channels, out_channels=None)

(self, in_channels)

(self, nf, emb_dim, ch_mult, res_blocks, img_size, attn_resolutions)

( self, img_size, nf, ch_mult, quantizer="nearest", res_blocks