SMAUG
Simulating Machine Learning Applications on gem5-Aladdin
Public Types | Public Member Functions | List of all members
smaug::PaddingOp< Backend > Class Template Reference

Pad a given tensor in any number of dimensions with arbitrary size. More...

#include <padding_op.h>

Detailed Description

template<typename Backend>
class smaug::PaddingOp< Backend >

Pad a given tensor in any number of dimensions with arbitrary size.

This has a software-based implementation.

Template Parameters
BackendThe Backend that sets Alignment.

Definition at line 61 of file backend.h.

Public Types

enum  { kInput, kNumInputs }
 
enum  { kOutput, kNumOutputs }
 

Public Member Functions

 PaddingOp (const std::string &name, Workspace *workspace)
 
void setPaddingSize (const RepeatedField< google::protobuf::int32 > &val)
 Set the paddingSize of the Tensor along each dimension. More...
 
void setPaddingSize (std::vector< int > const &val)
 
const std::vector< int > & getPaddingSize () const
 
void run () override
 
void createAllTensors () override
 
bool validate () override
 

Member Function Documentation

◆ setPaddingSize()

template<typename Backend >
void smaug::PaddingOp< Backend >::setPaddingSize ( const RepeatedField< google::protobuf::int32 > &  val)
inline

Set the paddingSize of the Tensor along each dimension.

The paddingSize is orgainized as <{dim0_begin, dim0_end, dim1_begin, dim1_end, ... >

Definition at line 34 of file padding_op.h.


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