SMAUG
Simulating Machine Learning Applications on gem5-Aladdin
Public Member Functions | Protected Member Functions | Protected Attributes | List of all members
smaug::SmvBatchNormOp Class Reference

SMV backend implementation of batch normalization. More...

#include <smv_batch_norm_op.h>

Inheritance diagram for smaug::SmvBatchNormOp:
smaug::BatchNormOp< SmvBackend >

Detailed Description

SMV backend implementation of batch normalization.

Elements are formatted and consumed in vectors of 8.

Definition at line 27 of file smv_batch_norm_op.h.

Public Member Functions

void tile () override
 
void run () override
 
- Public Member Functions inherited from smaug::BatchNormOp< SmvBackend >
 BatchNormOp (const std::string &name, Workspace *workspace)
 
void run () override
 
void run ()
 
TensorShape inferOutputShape () const
 
TensorShape inferWeightsShape () const
 
void createWeightsTensors ()
 
void createOutputTensors ()
 
void createAllTensors () override
 
int getNumParameters () const override
 
std::vector< TensorBase * > getParameterizableInputs () override
 
bool isSamplingSupported () const override
 
void setSamplingInfo (const SamplingInfo &_sampling) override
 

Protected Member Functions

void runNA (TiledTensor &inputs, TiledTensor &weights, TiledTensor &outputs)
 Post-FC tile dispatcher.
 
void runNHWC (TiledTensor &inputs, TiledTensor &weights, TiledTensor &outputs)
 Post-convolution tile dispatcher.
 

Protected Attributes

std::array< TiledTensor, 3 > tiledTensors
 
- Protected Attributes inherited from smaug::BatchNormOp< SmvBackend >
const std::string meanName
 
const std::string varianceName
 
const std::string gammaName
 
const std::string betaName
 
SamplingInfo sampling
 

Additional Inherited Members

- Public Types inherited from smaug::BatchNormOp< SmvBackend >
enum  
 
enum  
 
- Static Public Attributes inherited from smaug::BatchNormOp< SmvBackend >
static constexpr float kEpsilon
 

The documentation for this class was generated from the following files: