Simd Library Documentation.

Home | Release Notes | Download | Documentation | Issues | GitHub
Synet Framework

Functions for accelerating of inference of neural network in Synet Framework. More...

Modules

 Types
 Types used in Synet Framework.
 
 Activation functions
 Functions to acceleratе activation functions in Synet Framework.
 
 Conversion functions
 Functions to acceleratе conversion in Synet Framework.
 
 FP32 convolution framework
 A framework to accelerate FP32 convolution in Synet Framework.
 
 BF16 convolution framework
 A framework to accelerate BF16 convolution in Synet Framework.
 
 INT8 convolution framework
 A framework to accelerate INT8 convolution in Synet Framework.
 
 FP32 deconvolution framework
 A framework to accelerate FP32 deconvolution in Synet Framework.
 
 GridSampleLayer functions
 Functions to acceleratе GridSampleLayer in Synet Framework.
 
 InnerProductLayer functions
 Functions to acceleratе InnerProductLayer in Synet Framework.
 
 FP32 merged convolution frameworks
 A framework to accelerate FP32 merged convolution in Synet Framework.
 
 INT8 merged convolution frameworks
 A framework to accelerate INT8 merged convolution in Synet Framework.
 
 NormalizeLayer functions
 Functions to acceleratе NormalizeLayer in Synet Framework.
 
 PermuteLayer functions
 Functions to acceleratе PermuteLayer in Synet Framework.
 
 PoolingLayer functions
 Functions to acceleratе PoolingLayer in Synet Framework.
 
 ScaleLayer functions
 Functions to acceleratе layer scale in Synet Framework.
 
 Winograd functions
 Functions to acceleratе Winograd convolution algorithm in Synet Framework.
 
 Other functions
 Other accelerated functions used in Synet Framework.
 

Detailed Description

Functions for accelerating of inference of neural network in Synet Framework.