Functions — github.com/Tencent-Hunyuan/HunyuanPortrait

hub / github.com/Tencent-Hunyuan/HunyuanPortrait / functions

Functions229 in github.com/Tencent-Hunyuan/HunyuanPortrait

⨍ Functions 229◇ Types & classes 61

Methodforward

( self, hidden_states: torch.FloatTensor, res_hidden_states_tuple: Tuple[torch.FloatTe

src/models/condition/unet_3d_blocks.py:1265

Methodforward

( self, hidden_states: torch.FloatTensor, temb: Optional[torch.FloatTensor] = None,

src/models/condition/unet_3d_blocks.py:1382

Methodforward

( self, hidden_states: torch.FloatTensor, temb: Optional[torch.FloatTensor] = None,

src/models/condition/unet_3d_blocks.py:1548

Methodforward

( self, hidden_states: torch.FloatTensor, res_hidden_states_tuple: Tuple[torch.FloatTe

src/models/condition/unet_3d_blocks.py:1731

Methodforward

( self, hidden_states: torch.FloatTensor, res_hidden_states_tuple: Tuple[torch.FloatTe

src/models/condition/unet_3d_blocks.py:1892

Methodforward

( self, hidden_states: torch.FloatTensor, temb: Optional[torch.FloatTensor] = None,

src/models/condition/unet_3d_blocks.py:2073

Methodforward

( self, hidden_states: torch.FloatTensor, image_only_indicator: torch.FloatTensor,

src/models/condition/unet_3d_blocks.py:2185

Methodforward

( self, hidden_states: torch.FloatTensor, image_only_indicator: torch.FloatTensor,

src/models/condition/unet_3d_blocks.py:2236

Methodforward

( self, hidden_states: torch.FloatTensor, temb: Optional[torch.FloatTensor] = None,

src/models/condition/unet_3d_blocks.py:2309

Methodforward

( self, hidden_states: torch.FloatTensor, temb: Optional[torch.FloatTensor] = None,

src/models/condition/unet_3d_blocks.py:2407

Methodforward

( self, hidden_states: torch.FloatTensor, temb: Optional[torch.FloatTensor] = None,

src/models/condition/unet_3d_blocks.py:2517

Methodforward

( self, hidden_states: torch.FloatTensor, res_hidden_states_tuple: Tuple[torch.FloatTe

src/models/condition/unet_3d_blocks.py:2619

Methodforward

( self, hidden_states: torch.FloatTensor, res_hidden_states_tuple: Tuple[torch.FloatTe

src/models/condition/unet_3d_blocks.py:2728

Methodforward

(self, x)

src/models/condition/pose_guider.py:17

Methodget_intermediate_layers

( self, x: torch.Tensor, n: Union[int, Sequence] = 1, # Layers or n last layers to ta

src/models/dinov2/models/vision_transformer.py:350

Methodgetframes

(self)

src/dataset/utils.py:268

Methodguidance_scale1

(self)

src/pipelines/hunyuan_svd_pipeline.py:277

Methodguidance_scale2

(self)

src/pipelines/hunyuan_svd_pipeline.py:281

Methodindice_slice

(tensor, idx_list)

src/pipelines/hunyuan_svd_pipeline.py:577

Functioninit_weights_vit_timm

ViT weight initialization, original timm impl (for reproducibility)

src/models/dinov2/models/vision_transformer.py:392

Methodnum_timesteps

(self)

src/pipelines/hunyuan_svd_pipeline.py:292

Functionrun_inference

(image_input, video_input, steps, seed, frame_limit, fps, motion_bucket_

gradio_app.py:170

Methodset_default_attn_processor

Disables custom attention processors and sets the default attention implementation.

src/models/condition/unet_3d_svd_condition_ip.py:310

Methodspatial2time

(time_context)

src/models/condition/unet_3d_blocks.py:308

Methodstep_bk

Predict the sample at the previous timestep by reversing the SDE. Core function to propagate the diffusion process from the learned m

src/schedulers/scheduling_euler_discrete.py:203

Functiontensor2vid

(video: torch.Tensor, processor: VaeImageProcessor, output_type: str = "np")

src/pipelines/hunyuan_svd_pipeline.py:36

Functionvit_base

(patch_size=16, num_register_tokens=0, **kwargs)

src/models/dinov2/models/vision_transformer.py:414

Functionvit_giant2

Close to ViT-giant, with embed-dim 1536 and 24 heads => embed-dim per head 64

src/models/dinov2/models/vision_transformer.py:442

Functionvit_small

(patch_size=16, num_register_tokens=0, **kwargs)

src/models/dinov2/models/vision_transformer.py:400

← previous201–229 of 229, ranked by callers