Functions — github.com/antgroup/echomimic_v3

CodeHub MCPcopy Create free account

hub / github.com/antgroup/echomimic_v3 / functions

Functions301 in github.com/antgroup/echomimic_v3

⨍ Functions 301◇ Types & classes 66

↓ 28 callersMethodfrom_pretrained

(cls, pretrained_model_path, additional_kwargs={})

src/wan_vae.py:688

↓ 11 callersMethodset_timesteps

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

src/fm_solvers.py:226

↓ 10 callersMethod_sigma_to_alpha_sigma_t

src/fm_solvers.py:333

↓ 9 callersMethod__init__

(self, dim, out_dim, patch_size, eps=1e-6)

src/wan_transformer3d_audio_2512.py:972

↓ 9 callersMethod__init__

(self, dim, out_dim, patch_size, eps=1e-6)

src/wan_transformer3d_audio.py:922

↓ 8 callersMethod__init__

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

src/wan_vae.py:47

↓ 8 callersMethod_sigma_to_alpha_sigma_t

src/fm_solvers_unipc.py:272

↓ 8 callersFunctionfilter_kwargs

src/utils.py:12

↓ 8 callersFunctionrope_params

(max_seq_len, dim, theta=10000)

src/wan_transformer3d_audio_2512.py:495

↓ 8 callersFunctionrope_params

(max_seq_len, dim, theta=10000)

src/wan_transformer3d_audio.py:435

↓ 7 callersMethod__init__

(self, dim, mid_dim)

src/wan_image_encoder.py:98

↓ 7 callersFunctionattention

( q, k, v, q_lens=None, k_lens=None, dropout_p=0., softmax_scale=None, q_scale

src/wan_transformer3d_audio.py:165

↓ 7 callersFunctionget_image_to_video_latent3

(validation_image_start, validation_image_end, video_length, sample_size)

src/utils.py:377

↓ 6 callersMethod__init__

(self, dim, eps=1e-6)

src/wan_text_encoder.py:45

↓ 6 callersFunctionattention

( q, k, v, q_lens=None, k_lens=None, dropout_p=0., softmax_scale=None, q_scale

src/wan_transformer3d_audio_2512.py:241

↓ 6 callersFunctionhalf

src/wan_transformer3d_audio_2512.py:102

↓ 6 callersFunctionhalf

src/wan_transformer3d_audio.py:80

↓ 5 callersFunctionfp16_clamp

src/wan_text_encoder.py:14

↓ 4 callersMethodclear_cache

src/wan_vae.py:592

↓ 4 callersMethodenable_riflex

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

src/wan_transformer3d_audio.py:1128

↓ 4 callersMethodenable_teacache

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

src/wan_transformer3d_audio.py:1113

↓ 4 callersMethodencode

( self, extract_features, attention_mask=None, mask_time_indices=None,

src/wav2vec2.py:97

↓ 4 callersFunctionflash_attention

q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens

src/wan_transformer3d_audio.py:45

↓ 4 callersFunctionget_mask_coord

src/face_detect.py:8

↓ 4 callersFunctionget_teacache_coefficients

src/cache_utils.py:5

↓ 4 callersFunctionsave_videos_grid

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

src/utils.py:54

↓ 3 callersMethoddecode

(self, z: torch.Tensor, return_dict: bool = True)

src/wan_vae.py:680

↓ 3 callersFunctionflash_attention

q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens

src/wan_transformer3d_audio_2512.py:67

↓ 3 callersFunctionretrieve_timesteps

Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs

src/pipeline_wan_fun_inpaint_audio.py:45

↓ 3 callersFunctionretrieve_timesteps

Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs

src/pipeline_wan_fun_inpaint_audio_2512.py:48

↓ 3 callersMethodscale_model_input

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

src/fm_solvers.py:800

↓ 3 callersMethodstep

Predict the sample from the previous timestep by reversing the SDE. This function propagates the sample with the multistep DPMSolver.

src/fm_solvers.py:706

↓ 2 callersMethod__init__

(self, vocab_size=250002, max_seq_len=514, type_size=1,

src/wan_xlm_roberta.py:81

↓ 2 callersMethod_get_t5_prompt_embeds

( self, prompt: Union[str, List[str]] = None, num_videos_per_prompt: int = 1,

src/pipeline_wan_fun_inpaint_audio.py:193

↓ 2 callersMethod_get_t5_prompt_embeds

( self, prompt: Union[str, List[str]] = None, num_videos_per_prompt: int = 1,

src/pipeline_wan_fun_inpaint_audio_2512.py:197

↓ 2 callersMethod_threshold_sample

"Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the prediction of x_0 at t

src/fm_solvers.py:292

↓ 2 callersMethod_threshold_sample

"Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the prediction of x_0 at t

src/fm_solvers_unipc.py:230

↓ 2 callersFunctioncolor_transfer

Transfer color distribution from of sc, referred to dc. Args: sc (numpy.ndarray): input image to be transfered. dc (numpy.nd

src/utils.py:26

↓ 2 callersMethodcompute_rel_l1_distance

(prev: torch.Tensor, cur: torch.Tensor)

src/cache_utils.py:62

↓ 2 callersFunctioncount_conv3d

src/wan_vae.py:481

↓ 2 callersMethoddecode_latents

(self, latents: torch.Tensor)

src/pipeline_wan_fun_inpaint_audio.py:380

↓ 2 callersMethoddecode_latents

(self, latents: torch.Tensor)

src/pipeline_wan_fun_inpaint_audio_2512.py:379

↓ 2 callersMethodencode

(self, x, scale)

src/wan_vae.py:522

↓ 2 callersMethodencode_prompt

r""" Encodes the prompt into text encoder hidden states. Args: prompt (`str` or `List[str]`, *optional*):

src/pipeline_wan_fun_inpaint_audio.py:237

↓ 2 callersMethodforward

src/wan_vae.py:57

↓ 2 callersMethodforward

src/wan_image_encoder.py:108

↓ 2 callersFunctionget_mean_and_std

src/utils.py:38

↓ 2 callersFunctionget_sampling_sigmas

(sampling_steps, shift)

src/fm_solvers.py:22

↓ 2 callersMethodindex_for_timestep

(self, timestep, schedule_timesteps=None)

src/fm_solvers.py:679

↓ 2 callersMethodindex_for_timestep

(self, timestep, schedule_timesteps=None)

src/fm_solvers_unipc.py:628

↓ 2 callersFunctionlinear_interpolation

(features, seq_len)

src/wav2vec2.py:19

↓ 2 callersMethodreset

src/cache_utils.py:67

↓ 2 callersFunctionrope_apply

(x, grid_sizes, freqs)

src/wan_transformer3d_audio_2512.py:583

↓ 2 callersFunctionrope_apply

(x, grid_sizes, freqs)

src/wan_transformer3d_audio.py:523

↓ 1 callersMethod__init__

(self, in_dim, out_dim, kernel_size, stride, num_residual_blocks=1)

src/wan_camera_adapter.py:5

↓ 1 callersFunction_clip

(pretrained=False, pretrained_name=None, model_cls=XLMRobertaCLIP, return_transf

src/wan_image_encoder.py:436

↓ 1 callersMethod_decode

src/wan_vae.py:670

↓ 1 callersMethod_encode

(self, x: torch.Tensor)

src/wan_vae.py:650

↓ 1 callersMethod_init_step_index

Initialize the step_index counter for the scheduler.

src/fm_solvers.py:693

↓ 1 callersMethod_init_step_index

Initialize the step_index counter for the scheduler.

src/fm_solvers_unipc.py:643

↓ 1 callersMethod_norm

src/wan_transformer3d_audio_2512.py:628

↓ 1 callersMethod_norm

src/wan_transformer3d_audio.py:568

↓ 1 callersMethod_relative_position_bucket

(self, rel_pos)

src/wan_text_encoder.py:235

↓ 1 callersFunction_video_vae

Autoencoder3d adapted from Stable Diffusion 1.x, 2.x and XL.

src/wan_vae.py:602

↓ 1 callersFunctionaudio_mask_attention

( q, k, v, q_lens=None, k_lens=None, dropout_p=0., softmax_scale=None, q_scale

src/wan_transformer3d_audio_2512.py:432

↓ 1 callersFunctionaudio_mask_attention

( q, k, v, q_lens=None, k_lens=None, dropout_p=0., softmax_scale=None, q_scale

src/wan_transformer3d_audio.py:357

↓ 1 callersMethodcheck_inputs

( self, prompt, height, width, negative_prompt, callback_on_st

src/pipeline_wan_fun_inpaint_audio.py:406

↓ 1 callersMethodcheck_inputs

( self, prompt, height, width, negative_prompt, callback_on_st

src/pipeline_wan_fun_inpaint_audio_2512.py:400

↓ 1 callersFunctionclip_xlm_roberta_vit_h_14

( pretrained=False, pretrained_name='open-clip-xlm-roberta-large-vit-huge-14', **kwarg

src/wan_image_encoder.py:473

↓ 1 callersMethodconvert_model_output

Convert the model output to the corresponding type the DPMSolver/DPMSolver++ algorithm needs. DPM-Solver is designed to discretize an

src/fm_solvers.py:341

↓ 1 callersMethodconvert_model_output

r""" Convert the model output to the corresponding type the UniPC algorithm needs. Args: model_output (`torch.Tensor`):

src/fm_solvers_unipc.py:279

↓ 1 callersMethoddecode

(self, z, scale)

src/wan_vae.py:553

↓ 1 callersMethoddpm_solver_first_order_update

One step for the first-order DPMSolver (equivalent to DDIM). Args: model_output (`torch.Tensor`): The dir

src/fm_solvers.py:415

↓ 1 callersMethodenable_multi_gpus_inference

src/wan_transformer3d_audio.py:1155

↓ 1 callersMethodencode

( self, x: torch.Tensor, return_dict: bool = True )

src/wan_vae.py:659

↓ 1 callersMethodencode_prompt

r""" Encodes the prompt into text encoder hidden states. Args: prompt (`str` or `List[str]`, *optional*):

src/pipeline_wan_fun_inpaint_audio_2512.py:241

↓ 1 callersFunctionextract_audio_features

Extract audio features using Wav2Vec.

↓ 1 callersFunctionextract_audio_features

Extract audio features using Wav2Vec.

infer_preview.py:118

↓ 1 callersFunctionextract_audio_features

Extract audio features using Wav2Vec.

↓ 1 callersMethodforward

r""" Args: x(Tensor): Shape [B, L1, C] e(Tensor): Shape [B, C]

src/wan_transformer3d_audio_2512.py:987

↓ 1 callersMethodforward

r""" Args: x(Tensor): Shape [B, L1, C] e(Tensor): Shape [B, C]

src/wan_transformer3d_audio.py:937

↓ 1 callersFunctionget_1d_rotary_pos_embed_riflex

RIFLEx: Precompute the frequency tensor for complex exponentials (cis) with given dimensions. This function calculates a frequency tensor wi

src/wan_transformer3d_audio_2512.py:506

↓ 1 callersFunctionget_1d_rotary_pos_embed_riflex

RIFLEx: Precompute the frequency tensor for complex exponentials (cis) with given dimensions. This function calculates a frequency tensor wi

src/wan_transformer3d_audio.py:446

↓ 1 callersFunctionget_ai2v_audio

( audio, vae_scale, audio_window, device, dtype )

src/wan_transformer3d_audio_2512.py:45

↓ 1 callersFunctionget_audio_embed

(mel_input, wav2vec_feature_extractor, audio_encoder, video_length, sr=16000, fps=25, device='cpu')

infer_flash.py:134

↓ 1 callersFunctionget_audio_mask

ip_mask B， n_q

src/wan_transformer3d_audio.py:408

↓ 1 callersFunctionget_file_path

Helper function to find the file path with multiple extensions.

infer_preview.py:152

↓ 1 callersFunctionget_image_to_video_latent2

(validation_image_start, validation_image_end, video_length, sample_size)

src/utils.py:278

↓ 1 callersFunctionget_ip_mask

↓ 1 callersFunctionget_ip_mask

infer_preview.py:144

↓ 1 callersFunctionget_ip_mask

↓ 1 callersFunctionget_sample_size

(pil_img, sample_size)

infer_flash.py:110

↓ 1 callersFunctionget_sample_size

Calculate the sample size based on the input image dimensions.

↓ 1 callersFunctionget_sample_size

Calculate the sample size based on the input image dimensions.

infer_preview.py:127

↓ 1 callersFunctionget_sample_size

Calculate the sample size based on the input image dimensions.

↓ 1 callersFunctionload_wav2vec_models

Load Wav2Vec models for audio feature extraction.

↓ 1 callersFunctionload_wav2vec_models

Load Wav2Vec models for audio feature extraction.

infer_preview.py:110

↓ 1 callersFunctionload_wav2vec_models

Load Wav2Vec models for audio feature extraction.

↓ 1 callersFunctionloudness_norm

(audio_array, sr=16000, lufs=-23)

infer_flash.py:151

↓ 1 callersFunctionmain

infer_flash.py:159

next →1–100 of 301, ranked by callers