TI Deep Learning Product User Guide
sTIDL_BatchNormParams_t Struct Reference

Detailed Description

This structure define the parameters of Batch Norm layer in TIDL.

Data Fields

int32_t weights
 
int32_t bias
 
int32_t numChannels
 
int32_t biasQ
 
int32_t inDataQ
 
int32_t outDataQ
 
int32_t weightsQ
 
float32_tidl weightScale
 
float32_tidl biasScale
 
int32_t zeroWeightValue
 

Field Documentation

◆ weights

int32_t sTIDL_BatchNormParams_t::weights

Offset for kernel parameters

◆ bias

int32_t sTIDL_BatchNormParams_t::bias

Offset for bias parameters

◆ numChannels

int32_t sTIDL_BatchNormParams_t::numChannels

Number of channels channels to be processed

◆ biasQ

int32_t sTIDL_BatchNormParams_t::biasQ

Q value of Bias parameter

◆ inDataQ

int32_t sTIDL_BatchNormParams_t::inDataQ

Q value of the in data

◆ outDataQ

int32_t sTIDL_BatchNormParams_t::outDataQ

Q value expected for out data

◆ weightsQ

int32_t sTIDL_BatchNormParams_t::weightsQ

Q value of Kernel weights

◆ weightScale

float32_tidl sTIDL_BatchNormParams_t::weightScale

floating point scale for weight

◆ biasScale

float32_tidl sTIDL_BatchNormParams_t::biasScale

floating point scale for bias

◆ zeroWeightValue

int32_t sTIDL_BatchNormParams_t::zeroWeightValue

value of weights added for dynamic quantSytle