PadDescriptor Struct Reference

A PadDescriptor for the PadLayer. More...

#include <Descriptors.hpp>

Inheritance diagram for PadDescriptor:

Public Member Functions
	PadDescriptor ()
	PadDescriptor (const std::vector< std::pair< unsigned int, unsigned int >> &padList, const float &padValue=0, const PaddingMode &paddingMode=PaddingMode::Constant)
bool	operator== (const PadDescriptor &rhs) const
Public Member Functions inherited from BaseDescriptor
virtual bool	IsNull () const
virtual	~BaseDescriptor ()=default

Public Attributes
std::vector< std::pair< unsigned int, unsigned int > >	m_PadList
	Specifies the padding for input dimension. More...
float	m_PadValue
	Optional value to use for padding, defaults to 0. More...
PaddingMode	m_PaddingMode
	Specifies the Padding mode (Constant, Reflect or Symmetric) More...

Detailed Description

A PadDescriptor for the PadLayer.

Definition at line 1136 of file Descriptors.hpp.

Constructor & Destructor Documentation

PadDescriptor() [1/2]

PadDescriptor ( )

inline

Definition at line 1138 of file Descriptors.hpp.

                    : m_PadValue(0), m_PaddingMode(PaddingMode::Constant)
    {}

PadDescriptor() [2/2]

PadDescriptor ( const std::vector< std::pair< unsigned int, unsigned int >> &  padList, const float &  padValue = 0, const PaddingMode &  paddingMode = PaddingMode::Constant  )

inline

Definition at line 1141 of file Descriptors.hpp.

        : m_PadList(padList)
        , m_PadValue(padValue)
        , m_PaddingMode(paddingMode)
    {}

Member Function Documentation

operator==()

bool operator== ( const PadDescriptor &  rhs ) const

inline

Definition at line 1149 of file Descriptors.hpp.

References PadDescriptor::m_PaddingMode, PadDescriptor::m_PadList, and PadDescriptor::m_PadValue.

    {
        return m_PadList == rhs.m_PadList && m_PadValue == rhs.m_PadValue && m_PaddingMode == rhs.m_PaddingMode;
    }

Member Data Documentation

m_PaddingMode

PaddingMode m_PaddingMode

Specifies the Padding mode (Constant, Reflect or Symmetric)

Definition at line 1164 of file Descriptors.hpp.

Referenced by PadLayer::CreateWorkload(), RefPadWorkload::ExecuteAsync(), armnnSerializer::GetFlatBufferArgMinMaxFunction(), armnn::MirrorPad(), PadDescriptor::operator==(), TfLiteParserImpl::OutputShapeOfSqueeze(), and StringifyLayerParameters< PadDescriptor >::Serialize().

m_PadList

std::vector<std::pair<unsigned int, unsigned int> > m_PadList

Specifies the padding for input dimension.

First is the number of values to add before the tensor in the dimension. Second is the number of values to add after the tensor in the dimension. The number of pairs should match the number of dimensions in the input tensor.

Definition at line 1158 of file Descriptors.hpp.

Referenced by PadLayer::Clone(), ClPadWorkload::ClPadWorkload(), armnn_driver::ConvertPaddings(), PadLayer::CreateWorkload(), armnnSerializer::GetFlatBufferArgMinMaxFunction(), PadLayer::InferOutputShapes(), armnn::MirrorPad(), NeonPadWorkload::NeonPadWorkload(), PadDescriptor::operator==(), TfLiteParserImpl::OutputShapeOfSqueeze(), armnn::Pad(), StringifyLayerParameters< PadDescriptor >::Serialize(), and armnn::optimizations::pad_fold::TryFoldPadIntoLayer2d().

m_PadValue

float m_PadValue

Optional value to use for padding, defaults to 0.

Definition at line 1161 of file Descriptors.hpp.

Referenced by Converter::ConvertOperation(), armnnSerializer::GetFlatBufferArgMinMaxFunction(), PadDescriptor::operator==(), TfLiteParserImpl::OutputShapeOfSqueeze(), armnn::Pad(), StringifyLayerParameters< PadDescriptor >::Serialize(), and armnn::optimizations::pad_fold::TryFoldPadIntoLayer2d().

---

The documentation for this struct was generated from the following file:

• include/armnn/Descriptors.hpp