( self, tokenizer: AutoTokenizer, text_encoder: UMT5EncoderModel, vae: Autoenc

( self, tokenizer: T5Tokenizer, text_encoder: T5EncoderModel, vae: Autoencoder

( self, tokenizer: T5Tokenizer, text_encoder: T5EncoderModel, vae: Autoencoder

( self, in_channels: int, num_attention_heads: int, kernel_size: int, )

( self, in_channels: int, out_channels: int, num_attention_heads: Optional[int

( self, query_dim: int, added_kv_proj_dim: int, processor: "MochiAttnProcessor

(self)

( self, train_kv: bool = True, train_q_out: bool = True, hidden_size: Optional

(self)

(self)

(self)

(self)

(self)

(self, attention_op: Optional[Callable] = None)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

( self, hidden_size: int, cross_attention_dim: int, num_tokens=(4,), scale=1.0, device=None, dtype=Non

(self)

(self)

(self, attention_op: Optional[Callable] = None)

(self, attention_op: Optional[Callable] = None)

(self)

(self)

(self, partition_spec: Optional[Tuple[Optional[str], ...]] = None)

(self, partition_spec: Optional[Tuple[Optional[str], ...]] = None)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

( self, train_kv: bool = True, train_q_out: bool = False, hidden_size: Optiona

( self, train_kv: bool = True, train_q_out: bool = True, hidden_size: Optional

(self, slice_size: int)

(self, slice_size)

( self, f_channels: int, zq_channels: int, )

(self, hidden_size, cross_attention_dim=None, num_tokens=(4,), scale=1.0)

(self, hidden_size, cross_attention_dim=None, num_tokens=(4,), scale=1.0)

( self, hidden_size, cross_attention_dim=None, num_tokens=(4,), scale=

( self, hidden_size: int, ip_hidden_states_dim: int, head_dim: int, ti

(self)

(self)

(self)

(self)

(self)

(self)

(self)

( self, num_attention_heads: int = 30, attention_head_dim: int = 64, in_channe

(self)

( self, dim: int, time_freq_dim: int, time_proj_dim: int, text_embed_d

( self, attention_head_dim: int, patch_size: Tuple[int, int, int], max_seq_len

( self, dim: int, ffn_dim: int, num_heads: int, qk_norm: str = "rms_no

( self, patch_size: Tuple[int] = (1, 2, 2), num_attention_heads: int = 40, att

( self, height=224, width=224, patch_size=16, in_channels=3, e

(self, patch_size=2, in_channels=4, embed_dim=768, bias=True)

( self, patch_size: int = 2, patch_size_t: Optional[int] = None, in_channels:

( self, in_channels: int = 16, hidden_size: int = 2560, patch_size: int = 2,

(self, theta: int, axes_dim: List[int])

( self, in_channels: int, time_embed_dim: int, act_fn: str = "silu", o

( self, embedding_size: int = 256, scale: float = 1.0, set_W_to_weight=True, log=True, flip_sin_to_cos

(self, embed_dim: int, max_seq_length: int = 32)

( self, num_embed: int, height: int, width: int, embed_dim: int, )

(self, num_classes, hidden_size, dropout_prob)

( self, text_embed_dim: int = 1024, image_embed_dim: int = 768, cross_attentio

( self, image_embed_dim: int = 768, cross_attention_dim: int = 768, num_image_

(self, image_embed_dim=1024, cross_attention_dim=1024)

(self, image_embed_dim=1024, cross_attention_dim=1024, mult=1, num_tokens=1)

(self, num_classes, embedding_dim, class_dropout_prob=0.1)

(self, embedding_dim, pooled_projection_dim)

(self, embedding_dim, pooled_projection_dim)

(self, embedding_dim: int, condition_dim: int, pooled_projection_dim: int, timesteps_dim: int = 256)

(self, spacial_dim: int, embed_dim: int, num_heads: int, output_dim: int = None)

( self, embedding_dim, pooled_projection_dim=1024, seq_len=256, cross_

(self, hidden_size=4096, cross_attention_dim=2048, frequency_embedding_size=256)

( self, embedding_dim: int, pooled_projection_dim: int, text_embed_dim: int,

(self, encoder_dim: int, time_embed_dim: int, num_heads: int = 64)

(self, text_embed_dim: int = 768, image_embed_dim: int = 768, time_embed_dim: int = 1536)

(self, image_embed_dim: int = 768, time_embed_dim: int = 1536)

(self, image_embed_dim: int = 768, time_embed_dim: int = 1536)

(self, num_heads, embed_dim, dtype=None)

( self, num_attention_heads: int, embed_dim: int, output_dim: Optional[int] =

(self, positive_len, out_dim, feature_type="text-only", fourier_freqs=8)

(self, embedding_dim, size_emb_dim, use_additional_conditions: bool = False)

(self, in_features, hidden_size, out_features=None, act_fn="gelu_tanh")

( self, embed_dims: int = 768, dim_head: int = 64, heads: int = 16, ff

( self, embed_dims: int = 768, output_dims: int = 1024, hidden_dims: int = 128

( self, embed_dims: int = 768, output_dims: int = 768, hidden_dims: int = 1280

( self, hidden_dim: int = 1280, dim_head: int = 64, heads: int = 20, f

( self, embed_dim: int = 1152, output_dim: int = 2432, hidden_dim: int = 1280,

(self, IPAdapterImageProjectionLayers: Union[List[nn.Module], Tuple[nn.Module]])

(self, num_idx, uniform_sampling=False)

( self, in_channels, out_channels, factor_t, factor_s=1, )

( self, in_channels: int, out_channels: int, kernel_size: Union[int, Tuple[int