Types & classes — github.com/TaoRuijie/TalkNet-ASD

CodeHub MCPcopy Create free account

hub / github.com/TaoRuijie/TalkNet-ASD / types & classes

Types & classes23 in github.com/TaoRuijie/TalkNet-ASD

⨍ Functions 104◇ Types & classes 23

↓ 4 callersClassResNetLayer

A ResNet layer used to build the ResNet network. Architecture: --> conv-bn-relu -> conv -> + -> bn-relu -> conv-bn-relu -> conv -> + -

model/visualEncoder.py:11

↓ 4 callersClasstalkNet

↓ 3 callersClassL2Norm

model/faceDetector/s3fd/nets.py:8

↓ 3 callersClassattentionLayer

model/attentionLayer.py:6

↓ 1 callersClassDSConv1d

model/visualEncoder.py:132

↓ 1 callersClassDetect

model/faceDetector/s3fd/box_utils.py:129

↓ 1 callersClassGlobalLayerNorm

model/visualEncoder.py:83

↓ 1 callersClassPriorBox

model/faceDetector/s3fd/box_utils.py:176

↓ 1 callersClassResNet

An 18-layer ResNet architecture.

model/visualEncoder.py:57

↓ 1 callersClassS3FD

model/faceDetector/s3fd/__init__.py:17

↓ 1 callersClassS3FDNet

model/faceDetector/s3fd/nets.py:28

↓ 1 callersClassSELayer

model/audioEncoder.py:37

↓ 1 callersClassaudioEncoder

model/audioEncoder.py:54

↓ 1 callersClasslossA

↓ 1 callersClasslossAV

↓ 1 callersClasslossV

↓ 1 callersClasstalkNetModel

model/talkNetModel.py:8

↓ 1 callersClasstrain_loader

dataLoader.py:89

↓ 1 callersClassval_loader

dataLoader.py:124

↓ 1 callersClassvisualConv1D

model/visualEncoder.py:160

↓ 1 callersClassvisualFrontend

A visual feature extraction module. Generates a 512-dim feature vector per video frame. Architecture: A 3D convolution block followed by an

model/visualEncoder.py:100

↓ 1 callersClassvisualTCN

model/visualEncoder.py:148

ClassSEBasicBlock

model/audioEncoder.py:5