Functions — github.com/antgroup/echomimic

Methoddisable_riflex

(self)

src/wan_transformer3d_audio.py:1144

Methoddisable_teacache

(self)

src/wan_transformer3d_audio_2512.py:1178

Methoddisable_teacache

(self)

src/wan_transformer3d_audio.py:1125

Methodenable_multi_gpus_inference

(self,)

src/wan_transformer3d_audio_2512.py:1208

Methodenable_riflex

( self, k = 6, L_test = 66, L_test_scale = 4.886, )

src/wan_transformer3d_audio_2512.py:1181

Methodenable_teacache

( self, coefficients, num_steps: int, rel_l1_thresh: float, num_skip_s

src/wan_transformer3d_audio_2512.py:1166

Methodfeature_extract

( self, input_values, seq_len, )

src/wav2vec2.py:86

Methodfilter_kwargs

(cls, kwargs)

src/wan_text_encoder.py:308

Methodfilter_kwargs

(cls, kwargs)

src/wan_vae.py:689

Methodfilter_kwargs

(cls, kwargs)

src/wan_image_encoder.py:532

Methodforward

(self, audio_embeds, audio_embeds_vf)

src/wan_transformer3d_audio_2512.py:206

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

src/wan_transformer3d_audio_2512.py:621

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

src/wan_transformer3d_audio_2512.py:637

Methodforward

r""" Args: x(Tensor): Shape [B, L, num_heads, C / num_heads] seq_lens(Tensor): Shape [B] grid_sizes(Tensor

src/wan_transformer3d_audio_2512.py:670

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(Tensor): Shape [B, L2, C] context_lens(Tensor): Shape [B]

src/wan_transformer3d_audio_2512.py:705

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(Tensor): Shape [B, L2, C] context_lens(Tensor): Shape [B]

src/wan_transformer3d_audio_2512.py:747

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(List): Shape [B, L2, C], [B, L3, C] context_lens(Tensor): S

src/wan_transformer3d_audio_2512.py:810

Methodforward

r""" Args: x(Tensor): Shape [B, L, C] e(Tensor): Shape [B, 6, C] seq_lens(Tensor): Shape [B], length of ea

src/wan_transformer3d_audio_2512.py:926

Methodforward

(self, image_embeds)

src/wan_transformer3d_audio_2512.py:1008

Methodforward

r""" Forward pass through the diffusion model Args: x (List[Tensor]): List of input video tensors, each w

src/wan_transformer3d_audio_2512.py:1218

Methodforward

(self, x)

src/wan_camera_adapter.py:20

Methodforward

(self, x)

src/wan_camera_adapter.py:49

Methodforward

x: [B, L, C].

src/wan_xlm_roberta.py:27

Methodforward

(self, x, mask)

src/wan_xlm_roberta.py:66

Methodforward

ids: [B, L] of torch.LongTensor.

src/wan_xlm_roberta.py:118

Methodforward

(self, audio_embeds)

src/wan_transformer3d_audio.py:160

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

src/wan_transformer3d_audio.py:561

Methodforward

r""" Args: x(Tensor): Shape [B, L, C]

src/wan_transformer3d_audio.py:577

Methodforward

r""" Args: x(Tensor): Shape [B, L, num_heads, C / num_heads] seq_lens(Tensor): Shape [B] grid_sizes(Tensor

src/wan_transformer3d_audio.py:610

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(Tensor): Shape [B, L2, C] context_lens(Tensor): Shape [B]

src/wan_transformer3d_audio.py:645

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(Tensor): Shape [B, L2, C] context_lens(Tensor): Shape [B]

src/wan_transformer3d_audio.py:689

Methodforward

r""" Args: x(Tensor): Shape [B, L1, C] context(List): Shape [B, L2, C], [B, L3, C] context_lens(Tensor): S

src/wan_transformer3d_audio.py:760

Methodforward

r""" Args: x(Tensor): Shape [B, L, C] e(Tensor): Shape [B, 6, C] seq_lens(Tensor): Shape [B], length of ea

src/wan_transformer3d_audio.py:876

Methodforward

(self, image_embeds)

src/wan_transformer3d_audio.py:958

Methodforward

r""" Forward pass through the diffusion model Args: x (List[Tensor]): List of input video tensors, each w

src/wan_transformer3d_audio.py:1165

Methodforward

(self, x)

src/wan_text_encoder.py:39

Methodforward

(self, x)

src/wan_text_encoder.py:51

Methodforward

x: [B, L1, C]. context: [B, L2, C] or None. mask: [B, L2] or [B, L1, L2] or None.

src/wan_text_encoder.py:75

Methodforward

(self, x)

src/wan_text_encoder.py:125

Methodforward

(self, x, mask=None, pos_bias=None)

src/wan_text_encoder.py:158

Methodforward

(self, x, mask=None, encoder_states=None, enco

src/wan_text_encoder.py:193

Methodforward

(self, lq, lk)

src/wan_text_encoder.py:219

Methodforward

( self, input_ids: Optional[torch.LongTensor] = None, attention_mask: Optional[torch.

src/wan_text_encoder.py:291

Methodforward

( self, input_values, seq_len, attention_mask=None, mask_time_indices=

src/wav2vec2.py:32

Methodforward

(self, x, cache_x=None)

src/wan_vae.py:34

Methodforward

Fix bfloat16 support for nearest neighbor interpolation.

src/wan_vae.py:65

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

src/wan_vae.py:107

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

src/wan_vae.py:208

Methodforward

(self, x)

src/wan_vae.py:246

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

src/wan_vae.py:324

Methodforward

(self, x, feat_cache=None, feat_idx=[0])

src/wan_vae.py:429

Methodforward

(self, x)

src/wan_vae.py:516

Methodforward

(self, x)

src/wan_image_encoder.py:45

Methodforward

(self, x)

src/wan_image_encoder.py:51

Methodforward

x: [B, L, C].

src/wan_image_encoder.py:76

Methodforward

(self, x)

src/wan_image_encoder.py:148

Methodforward

x: [B, L, C].

src/wan_image_encoder.py:188

Methodforward

(self, x, interpolation=False, use_31_block=False)

src/wan_image_encoder.py:281

Methodforward

(self, ids)

src/wan_image_encoder.py:317

Methodforward

imgs: [B, 3, H, W] of torch.float32. - mean: [0.48145466, 0.4578275, 0.40821073] - std: [0.26862954, 0.2613025

src/wan_image_encoder.py:408

Methodforward

(self, videos)

src/wan_image_encoder.py:513

Methodfrom_pretrained

( cls, pretrained_model_path, subfolder=None, transformer_additional_kwargs={}, low_cpu_mem_us

src/wan_transformer3d_audio_2512.py:1504

Methodfrom_pretrained

( cls, pretrained_model_path, subfolder=None, transformer_additional_kwargs={}, low_cpu_mem_us

src/wan_transformer3d_audio.py:1485

Methodfrom_pretrained

(cls, pretrained_model_path, additional_kwargs={}, low_cpu_mem_usage=False, torch_dtype=torch.bfloat16)

src/wan_text_encoder.py:307

Methodfrom_pretrained

(cls, pretrained_model_path, transformer_additional_kwargs={})

src/wan_image_encoder.py:531

Functiongenerate

( image, audio, prompt, negative_prompt, partial_video_length, guidance_scale, aud

app_mm.py:241

Functiongenerate

( image, audio, prompt, negative_prompt, seed_param )

app.py:222

Functionget_image_latent

(ref_image=None, sample_size=None)

src/utils.py:263

Functionget_image_to_video_latent

(validation_image_start, validation_image_end, video_length, sample_size)

src/utils.py:106

Functionget_ip_mask

(coords)

infer_flash.py:126

Functionget_mask_from_lengths

(lengths, max_len=None)

src/wav2vec2.py:8

Functionget_resize_crop_region_for_grid

(src, tgt_width, tgt_height)

src/wan_transformer3d_audio_2512.py:565

Functionget_resize_crop_region_for_grid

(src, tgt_width, tgt_height)

src/wan_transformer3d_audio.py:505

Functionget_video_to_video_latent

(input_video_path, video_length, sample_size, fps=None, validation_video_mask=None, ref_image=None)

src/utils.py:202

Functionget_width_and_height_from_image_and_base_resolution

(image, base_resolution)

src/utils.py:18

Methodguidance_scale

(self)

src/pipeline_wan_fun_inpaint_audio.py:458

Methodguidance_scale

(self)

src/pipeline_wan_fun_inpaint_audio_2512.py:452

Methodinit_weight

(self, conv)

src/wan_vae.py:168

Methodinit_weight2

(self, conv)

src/wan_vae.py:180

Functioninit_weights

(m)

src/wan_text_encoder.py:21

Methodinit_weights

r""" Initialize model parameters using Xavier initialization.

src/wan_transformer3d_audio_2512.py:1479

Methodinit_weights

r""" Initialize model parameters using Xavier initialization.

src/wan_transformer3d_audio.py:1460

Methodinterrupt

(self)

src/pipeline_wan_fun_inpaint_audio.py:474

Methodinterrupt

(self)

src/pipeline_wan_fun_inpaint_audio_2512.py:464

Methodnum_timesteps

(self)

src/pipeline_wan_fun_inpaint_audio.py:466

Methodnum_timesteps

(self)

src/pipeline_wan_fun_inpaint_audio_2512.py:456

Methodparam_groups

(self)

src/wan_image_encoder.py:420

Methodqkv_fn

(x)

src/wan_transformer3d_audio_2512.py:681

Methodqkv_fn

(x)

src/wan_transformer3d_audio.py:621

Functionretrieve_timesteps

( scheduler, num_inference_steps=None, device=None, timesteps=None, sigmas=None, **kwa

src/fm_solvers.py:29

Methodsample

(self, imgs, deterministic=False)

src/wan_vae.py:585

Functionsave_videos_grid2

(videos: torch.Tensor, path: str, rescale=False, n_rows=6, fps=12, imageio_backend=True, color_transfer_post_p

src/utils.py:80

Methodscale_model_input

Ensures interchangeability with schedulers that need to scale the denoising model input depending on the current timestep. A

src/fm_solvers_unipc.py:741

Methodset_begin_index

Sets the begin index for the scheduler. This function should be run from pipeline before the inference. Args: begin_index

src/fm_solvers.py:216

Methodset_begin_index

Sets the begin index for the scheduler. This function should be run from pipeline before the inference. Args: begin_inde

src/fm_solvers_unipc.py:149

Methodset_timesteps

Sets the discrete timesteps used for the diffusion chain (to be run before inference). Args: num_inference_steps (`int`):

src/fm_solvers_unipc.py:160

Functionsplit_audio_sequence

Map the audio feature sequence to corresponding latent frame slices. Args: audio_proj_length (int): The total length of

src/wan_transformer3d_audio_2512.py:290

Functionsplit_tensor_with_padding

Split the input tensor into subsequences based on index ranges, and apply right-side zero-padding if the range exceeds the input boundaries.

src/wan_transformer3d_audio_2512.py:331

Methodstep_index

The index counter for current timestep. It will increase 1 after each scheduler step.

src/fm_solvers.py:202

Methodstep_index

The index counter for current timestep. It will increase 1 after each scheduler step.

src/fm_solvers_unipc.py:135

Functions301 in github.com/antgroup/echomimic_v3