tensor container that does memory allocation and resize like STL, use it to save the lines of FreeSpace in class. Do not abuse it, efficiency can come from pre-allocation and no re-allocation More...

#include <tensor_container.h>

Inheritance diagram for mshadow::TensorContainer< Device, dimension >:

Public Member Functions
	TensorContainer (bool pad=MSHADOW_ALLOC_PAD)
	constructor More...

	TensorContainer (const Shape< dimension > &shape)
	constructor More...

	TensorContainer (const Shape< dimension > &shape, real_t initv)
	constructor More...

void	Resize (const Shape< dimension > &shape)
	resize the container to given shape, content is NOT preserved More...

void	Resize (const Shape< dimension > &shape, real_t initv)
	resize the container to given shape, and initialize, content is NOT preserved More...

void	set_pad (bool pad)
	set whether padding is allowed in tensor

template<typename TStream >
void	SaveBinary (TStream &fo) const
	save by binary format More...

template<typename TStream >
void	LoadBinary (TStream &fi)
	load by binary format, a temp Tensor<cpu,dim> storage will be allocated More...

Tensor< Device, dimension > &	operator= (real_t s)

template<typename E >
Tensor< Device, dimension > &	operator= (const expr::Exp< E, expr::type::kMapper > &exp)

template<typename E >
Tensor< Device, dimension > &	operator= (const expr::Exp< E, expr::type::kComplex > &exp)

Public Member Functions inherited from mshadow::Tensor< Device, dimension >
MSHADOW_XINLINE	Tensor (void)
	default constructor

MSHADOW_XINLINE	Tensor (const Shape< dimension > &shape)
	constructor from shape

MSHADOW_XINLINE	Tensor (real_t *dptr, const Shape< dimension > &shape)
	constructor from data pointer and shape

MSHADOW_XINLINE Tensor< Device, 2 >	FlatTo2D (void) const
	flatten the tensor to 2 dimension, collapse the higher dimensions together More...

MSHADOW_XINLINE Tensor< Device, kSubdim >	operator[] (index_t idx) const
	get a element of dimension - 1 More...

MSHADOW_XINLINE Tensor< Device, dimension >	Slice (index_t begin, index_t end) const
	slice the tensor in highest dimension [begin,end) More...

Tensor< Device, dimension > &	operator= (real_t s)
	functions to fit expression template

template<typename E >
Tensor< Device, dimension > &	operator= (const expr::Exp< E, expr::type::kMapper > &exp)
	functions to fit expression template

template<typename E >
Tensor< Device, dimension > &	operator= (const expr::Exp< E, expr::type::kComplex > &exp)
	functions to fit expression template

Public Member Functions inherited from mshadow::expr::ContainerExp< Tensor< Device, dimension > >
const TransposeExp< Tensor < Device, dimension > >	T (void) const
	transpose of a matrix More...

Tensor< Device, dimension > &	operator+= (real_t s)
	operator overload

Tensor< Device, dimension > &	operator+= (const Exp< E, etype > &exp)
	implementation of operator+=

Tensor< Device, dimension > &	operator-= (real_t s)
	operator overload

Tensor< Device, dimension > &	operator-= (const Exp< E, etype > &exp)
	implementation of operator-=

Tensor< Device, dimension > &	operator*= (real_t s)
	operator overload

Tensor< Device, dimension > &	operator*= (const Exp< E, etype > &exp)
	implementation of operator*=

Tensor< Device, dimension > &	operator/= (real_t s)
	operator overload

Tensor< Device, dimension > &	operator/= (const Exp< E, etype > &exp)
	implementation of operator/=

Tensor< Device, dimension > &	__assign (real_t s)
	operator overload

Tensor< Device, dimension > &	__assign (const Exp< E, type::kMapper > &exp)
	implementation of operator=, note that we can not define container = container

Tensor< Device, dimension > &	__assign (const Exp< E, type::kComplex > &exp)
	implementation of operator=, note that we can not define container = container

Public Member Functions inherited from mshadow::expr::Exp< Tensor< Device, dimension >, type::kContainer >
const Tensor< Device, dimension > &	self (void) const

Tensor< Device, dimension > &	refself (void)

Additional Inherited Members
Public Attributes inherited from mshadow::Tensor< Device, dimension >
real_t *	dptr
	pointer to the data

Shape< dimension >	shape
	shape of the tensor

Static Public Attributes inherited from mshadow::Tensor< Device, dimension >
static const bool	kDevCPU = Device::kDevCPU
	whether current type lies in cpu

static const int	kSubdim = dimension - 1
	dimension of subtype

Detailed Description

template<typename Device, int dimension>
class mshadow::TensorContainer< Device, dimension >

tensor container that does memory allocation and resize like STL, use it to save the lines of FreeSpace in class. Do not abuse it, efficiency can come from pre-allocation and no re-allocation

Template Parameters

Device	which device the tensor is on
dimension	dimension of the tensor

Constructor & Destructor Documentation

template<typename Device, int dimension>

mshadow::TensorContainer< Device, dimension >::TensorContainer ( bool pad = MSHADOW_ALLOC_PAD )

inline

constructor

Parameters

pad	whether use padding alignment in space allocation

template<typename Device, int dimension>

mshadow::TensorContainer< Device, dimension >::TensorContainer ( const Shape< dimension > & shape )

inline

constructor

Parameters

shape intial shape

template<typename Device, int dimension>

mshadow::TensorContainer< Device, dimension >::TensorContainer	(	const Shape< dimension > &	shape,
		real_t	initv
	)

inline

constructor

Parameters

shape	intial shape
initv	intial value

Member Function Documentation

template<typename Device, int dimension>

template<typename TStream >

void mshadow::TensorContainer< Device, dimension >::LoadBinary ( TStream & fi )

inline

load by binary format, a temp Tensor<cpu,dim> storage will be allocated

Parameters

fi	input binary stream

Template Parameters

TStream type of stream, need to support Read, Write, one example is utils::IStream.

template<typename Device, int dimension>

void mshadow::TensorContainer< Device, dimension >::Resize ( const Shape< dimension > & shape )

inline

resize the container to given shape, content is NOT preserved

Parameters

shape target shape

template<typename Device, int dimension>

void mshadow::TensorContainer< Device, dimension >::Resize	(	const Shape< dimension > &	shape,
		real_t	initv
	)

inline

resize the container to given shape, and initialize, content is NOT preserved

Parameters

shape	target shape
initv	initialization value

template<typename Device, int dimension>

template<typename TStream >

void mshadow::TensorContainer< Device, dimension >::SaveBinary ( TStream & fo ) const

inline

save by binary format

Parameters

fo	output binary stream

Template Parameters

TStream type of stream, need to support Read, Write, one example is utils::IStream.

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

/home/wangwei/program/asf/incubator-singa/include/mshadow/tensor_container.h

Public Member Functions

Additional Inherited Members

Detailed Description

template<typename Device, int dimension> class mshadow::TensorContainer< Device, dimension >

Constructor & Destructor Documentation

Member Function Documentation

template<typename Device, int dimension>
class mshadow::TensorContainer< Device, dimension >