tensor blob class that can be used to hold tensor of any dimension, any device and any data type, This is a weak type that can be used to transfer data through interface TBlob itself do not involve any arithmentic operations, but it can be converted to tensor of fixed dimension for further operations More...

#include <tensor_blob.h>

Collaboration diagram for mxnet::TBlob:

Public Member Functions
	TBlob (void)
	default constructor, default copy assign will work More...

template<typename DType >
	TBlob (DType *dptr, const mxnet::TShape &shape, int dev_mask, int dev_id=-1)
	constructor that construct TBlob from contiguous memory More...

	TBlob (void *dptr, const mxnet::TShape &shape, int dev_mask, int type_flag, int dev_id=-1)
	constructor that construct TBlob from contiguous memory More...

	TBlob (const DLTensor &dltensor)
	constructor that construct TBlob from DLTensor More...

template<typename Device , int dim, typename DType >
	TBlob (const mshadow::Tensor< Device, dim, DType > &src)
	constructor from tensor More...

template<typename Device , int dim, typename DType >
TBlob &	operator= (const mshadow::Tensor< Device, dim, DType > &src)
	assignment from tensor More...

bool	CheckContiguous (void) const

TBlob	reshape (const mxnet::TShape &shape) const
	reshape to shape More...

template<typename Device , typename DType >
mshadow::Tensor< Device, 2, DType >	FlatTo2D (mshadow::Stream< Device > *stream=NULL) const
	flatten the tensor to 2 dimension, collapse the higher dimensions together More...

template<typename Device , typename DType >
mshadow::Tensor< Device, 1, DType >	FlatTo1D (mshadow::Stream< Device > *stream=NULL) const
	flatten the tensor to 1 dimension, collapse all the dimensions together. More...

int	ndim (void) const
	return number of dimension of the tensor inside More...

index_t	size (index_t idx) const
	return size of i-th dimension, start counting from highest dimension. return type needs to be a signed integer. More...

size_t	Size (void) const
	total number of elements in the tensor More...

template<typename DType >
DType *	dptr () const
	get pointer in dtype More...

int	dev_mask () const
	device mask of the corresponding device More...

int	dev_id () const
	device index of the corresponding device More...

const DLTensor &	dltensor () const
	return the corresponding DLTensor More...

template<typename Device , int dim, typename DType >
mshadow::Tensor< Device, dim, DType >	get (mshadow::Stream< Device > *stream=NULL) const
	fetch the tensor, with respect to specific dimension if dim do not match the stored dimension, an error will be issued More...

template<typename Device , int dim, typename DType >
mshadow::Tensor< Device, dim, DType >	get_with_shape (const mshadow::Shape< dim > &shape, mshadow::Stream< Device > *stream=NULL) const
	fetch a tensor in given shape If size do not match the stored size, an error will be issued More...

template<typename Device , typename DType >
mshadow::Tensor< Device, 3, DType >	FlatTo3D (int axis, mshadow::Stream< Device > *stream=NULL) const
	flatten the tensor to 3 dimension, collapse the dimension before and after specified axis. More...

template<typename Device , typename DType >
mshadow::Tensor< Device, 3, DType >	FlatTo3D (int axis_begin, int axis_end, mshadow::Stream< Device > *stream=NULL) const
	flatten the tensor to 3 dimension, collapse the dimension: [0, axis_begin), [axis_begin, axis_end], (axis_end, ndim). More...

template<typename Device , int dim, typename DType >
mshadow::Tensor< Device, dim, DType >	FlatToKD (mshadow::Stream< Device > *stream=NULL) const
	flatten the tensor to specified number of dimensions, collapse the highest dimensions or pad with higher dimensions More...

Public Attributes
void *	dptr_
	pointer to the data More...

mxnet::TShape	shape_
	shape of the tensor More...

int	type_flag_
	type flag of the tensor blob More...

Friends
class	NDArray

Detailed Description

tensor blob class that can be used to hold tensor of any dimension, any device and any data type, This is a weak type that can be used to transfer data through interface TBlob itself do not involve any arithmentic operations, but it can be converted to tensor of fixed dimension for further operations

Like tensor, this data structure is like a pointer class and do not implicit allocated, de-allocate space. This data structure can be helpful to hold tensors of different dimensions and wait for further processing

Constructor & Destructor Documentation

mxnet::TBlob::TBlob ( void )

inline

default constructor, default copy assign will work

template<typename DType >

mxnet::TBlob::TBlob	(	DType *	dptr,
		const mxnet::TShape &	shape,
		int	dev_mask,
		int	dev_id = `-1`
	)

inline

constructor that construct TBlob from contiguous memory

Parameters

dptr	the pointer to the memory
shape	the shape of the data
dev_mask	the device mask, can be cpu::kDevMask or gpu::kDevMask
dev_id	the device id

mxnet::TBlob::TBlob	(	void *	dptr,
		const mxnet::TShape &	shape,
		int	dev_mask,
		int	type_flag,
		int	dev_id = `-1`
	)

inline

constructor that construct TBlob from contiguous memory

Parameters

dptr	the pointer to the memory
shape	the shape of the data
dev_mask	the device mask, can be cpu::kDevMask or gpu::kDevMask
type_flag	the type flag. Can be one of enum mshadow::dtype
dev_id	the device id

mxnet::TBlob::TBlob ( const DLTensor & dltensor )

inlineexplicit

constructor that construct TBlob from DLTensor

Parameters

DLTensor Object

template<typename Device , int dim, typename DType >

mxnet::TBlob::TBlob ( const mshadow::Tensor< Device, dim, DType > & src )

inline

constructor from tensor

Parameters

src	source tensor

Template Parameters

Device	which device the tensor is on
dim	tensor dimension
DType	the type of elements in the tensor

Member Function Documentation

bool mxnet::TBlob::CheckContiguous ( void ) const

inline

Returns: whether the tensor's memory is continuous

int mxnet::TBlob::dev_id ( ) const

inline

device index of the corresponding device

int mxnet::TBlob::dev_mask ( ) const

inline

device mask of the corresponding device

const DLTensor& mxnet::TBlob::dltensor ( ) const

inline

return the corresponding DLTensor

Returns: the address of internal DLTensor

template<typename DType >

DType* mxnet::TBlob::dptr ( ) const

inline

get pointer in dtype

template<typename Device , typename DType >

mshadow::Tensor<Device, 1, DType> mxnet::TBlob::FlatTo1D ( mshadow::Stream< Device > * stream = NULL ) const

inline

flatten the tensor to 1 dimension, collapse all the dimensions together.

Parameters

stream the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
DType	the type of elements in the tensor

Returns: tensor after flatten

template<typename Device , typename DType >

mshadow::Tensor<Device, 2, DType> mxnet::TBlob::FlatTo2D ( mshadow::Stream< Device > * stream = NULL ) const

inline

flatten the tensor to 2 dimension, collapse the higher dimensions together

Parameters

stream the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
DType	the type of elements in the tensor

Returns: tensor after flatten

template<typename Device , typename DType >

mshadow::Tensor<Device, 3, DType> mxnet::TBlob::FlatTo3D	(	int	axis,
		mshadow::Stream< Device > *	stream = `NULL`
	)		const

inline

flatten the tensor to 3 dimension, collapse the dimension before and after specified axis.

Parameters

axis	The axis specified.
stream	the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
DType	the type of elements in the tensor

Returns: tensor after flatten

template<typename Device , typename DType >

mshadow::Tensor<Device, 3, DType> mxnet::TBlob::FlatTo3D	(	int	axis_begin,
		int	axis_end,
		mshadow::Stream< Device > *	stream = `NULL`
	)		const

inline

flatten the tensor to 3 dimension, collapse the dimension: [0, axis_begin), [axis_begin, axis_end], (axis_end, ndim).

Parameters

axis_begin	The beginning axis specified.
axis_end	The ending axis specified.
stream	the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
DType	the type of elements in the tensor

Returns: tensor after flatten

template<typename Device , int dim, typename DType >

mshadow::Tensor<Device, dim, DType> mxnet::TBlob::FlatToKD ( mshadow::Stream< Device > * stream = NULL ) const

inline

flatten the tensor to specified number of dimensions, collapse the highest dimensions or pad with higher dimensions

Parameters

stream the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
dim	desired number of dimensions of returned tensor
DType	the type of elements in the tensor

Returns: tensor after flatten

template<typename Device , int dim, typename DType >

mshadow::Tensor<Device, dim, DType> mxnet::TBlob::get ( mshadow::Stream< Device > * stream = NULL ) const

inline

fetch the tensor, with respect to specific dimension if dim do not match the stored dimension, an error will be issued

Returns: the tensor requested

Parameters

stream the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
dim	dimension of the tensor
DType	the type of elements in the tensor

template<typename Device , int dim, typename DType >

mshadow::Tensor<Device, dim, DType> mxnet::TBlob::get_with_shape	(	const mshadow::Shape< dim > &	shape,
		mshadow::Stream< Device > *	stream = `NULL`
	)		const

inline

fetch a tensor in given shape If size do not match the stored size, an error will be issued

Returns: the tensor requested

Parameters

shape	the shape required
stream	the possible stream target tensor should reside on

Template Parameters

Device	which device the tensor is on
dim	dimension of the tensor
DType	the type of elements in the tensor

int mxnet::TBlob::ndim ( void ) const

inline

return number of dimension of the tensor inside

template<typename Device , int dim, typename DType >

TBlob& mxnet::TBlob::operator= ( const mshadow::Tensor< Device, dim, DType > & src )

inline

assignment from tensor

Parameters

src	source tensor

Template Parameters

Device	which device the tensor is on
dim	tensor dimension
DType	the type of elements in the tensor

Returns: reference of self

TBlob mxnet::TBlob::reshape ( const mxnet::TShape & shape ) const

inline

reshape to shape

Parameters

shape desired shape

Returns: reshaped blob

index_t mxnet::TBlob::size ( index_t idx ) const

inline

return size of i-th dimension, start counting from highest dimension. return type needs to be a signed integer.

Parameters

idx	the dimension count from the highest dimensin

Returns: the size. -1 means unknown size to support zero-size tensor.

size_t mxnet::TBlob::Size ( void ) const

inline

total number of elements in the tensor

Friends And Related Function Documentation

friend class NDArray

friend

Member Data Documentation

void* mxnet::TBlob::dptr_

pointer to the data

mxnet::TShape mxnet::TBlob::shape_

shape of the tensor

int mxnet::TBlob::type_flag_

type flag of the tensor blob

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

include/mxnet/tensor_blob.h

Public Member Functions

Public Attributes

Friends

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Friends And Related Function Documentation

Member Data Documentation