ndarray interface More...

#include <ndarray.h>

Collaboration diagram for mxnet::NDArray:

Public Member Functions
	NDArray ()
	default constructor More...

	NDArray (const TShape &shape, Context ctx, bool delay_alloc=false, int dtype=mshadow::default_type_flag)
	constructs a new dynamic NDArray More...

	NDArray (const NDArrayStorageType stype, const TShape &shape, Context ctx, bool delay_alloc=true, int dtype=mshadow::default_type_flag, std::vector< int > aux_types={}, std::vector< TShape > aux_shapes={}, TShape storage_shape=TShape(mshadow::Shape1(0)))
	constructor for NDArray with storage type More...

	NDArray (const TBlob &data, int dev_id)
	constructing a static NDArray that shares data with TBlob Use with caution: allocate ONLY ONE NDArray for each TBlob, make sure the memory region is available through out the life of NDArray More...

	NDArray (const NDArrayStorageType stype, const TShape &shape, const TBlob &data, const std::vector< TBlob > &aux_data, int dev_id)
	constructing a static NDArray of non-default storage that shares data with TBlob Use with caution: allocate ONLY ONE NDArray for each TBlob, make sure the memory region is available through out the life of NDArray More...

const TShape &	shape () const

const TShape &	storage_shape () const

const TShape &	aux_shape (size_t index) const
	get the shape of aux_data(index) More...

const std::vector< TShape > &	aux_shapes () const

const std::vector< int > &	aux_types () const

void	set_aux_shape (size_t index, const TShape &shape) const
	For a sparse operation on a csr matrix for example, the size of the column index array is an estimated value in the beginning for allocating enough capacity for the final result. After the operation is done, the exact size of the shape is known and need to be reset using this function. More...

const TBlob &	data () const

NDArray	grad () const

TBlob	aux_data (size_t i) const

Context	ctx () const

int	dtype () const

int	aux_type (size_t i) const

NDArrayStorageType	storage_type () const

bool	is_none () const

bool	fresh_out_grad () const

void	set_fresh_out_grad (bool state) const

bool	storage_initialized () const

void	WaitToRead () const
	Block until all the pending write operations with respect to current NDArray are finished, and read can be performed. More...

void	WaitToWrite () const
	Block until all the pending read/write operations with respect to current NDArray are finished, and write can be performed. More...

Engine::VarHandle	var () const

void	Save (dmlc::Stream *strm) const
	save the content into binary stream More...

bool	LegacyLoad (dmlc::Stream *strm, const uint32_t magic)
	load ndarrays before supporting sparse ndarrays More...

bool	Load (dmlc::Stream *strm)
	load the content from binary stream More...

NDArray &	operator= (real_t scalar)
	set all the elements in ndarray to be scalar More...

NDArray &	operator+= (const NDArray &src)
	elementwise add to current space this mutate the current NDArray More...

NDArray &	operator+= (const real_t &src)
	elementwise add to current space this mutate the current NDArray More...

NDArray &	operator-= (const NDArray &src)
	elementwise subtract from current ndarray this mutate the current NDArray More...

NDArray &	operator-= (const real_t &src)
	elementwise subtract from current ndarray this mutate the current NDArray More...

NDArray &	operator*= (const NDArray &src)
	elementwise multiplication to current ndarray this mutate the current NDArray More...

NDArray &	operator*= (const real_t &src)
	elementwise multiplication to current ndarray this mutate the current NDArray More...

NDArray &	operator/= (const NDArray &src)
	elementwise division from current ndarray this mutate the current NDArray More...

NDArray &	operator/= (const real_t &src)
	elementwise division from current ndarray this mutate the current NDArray More...

NDArray	Copy (Context ctx) const
	return a new copy this NDArray More...

void	SyncCopyFromCPU (const void *data, size_t size) const
	Do a synchronize copy from a continugous CPU memory region. More...

void	SyncCopyFromNDArray (const NDArray &src, int i=-1, int j=-1)
	Copy from src.data()/aux_data(i) to this->data()/aux_data(j) More...

void	SyncCopyToCPU (void *data, size_t size) const
	Do a synchronize copy to a continugous CPU memory region. More...

NDArray	Slice (index_t begin, index_t end) const
	Slice a NDArray. More...

NDArray	SliceWithRecord (index_t begin, index_t end)
	Slice a NDArray. Supports recording with autograd. More...

NDArray	At (index_t idx) const
	Index a NDArray. More...

NDArray	AtWithRecord (index_t idx)
	Index a NDArray. More...

NDArray	aux_ndarray (size_t i) const
	Generate a deep copy of aux_data(i) returned as a default storage type NDArray. More...

NDArray	data_ndarray () const
	Generate a deep copy of data() returned as a default storage type NDArray. More...

NDArray	AsArray (const TShape &shape, int dtype) const
	Create a NDArray that shares memory with current one The new array must have smaller memory size than the current array. More...

NDArray	Reshape (const TShape &shape) const
	Get an reshaped NDArray. More...

NDArray	ReshapeWithRecord (const TShape &shape)
	Get an reshaped NDArray. Supports autograd recording. More...

NDArray	Detach () const
	Return a copy of this NDArray without autograd history. More...

nnvm::Symbol	get_autograd_symbol () const

void	CheckAndAlloc () const
	Allocate the space if it is delayed allocated. This is an internal function used by system that normal user should not use. More...

void	ReshapeAndAlloc (const TShape &shape)
	Allocate the space if the allocation has been delayed or the requested size is bigger than the available one. This function can only be called by ndarray of default storage type and effectively changes the ndarray's shape_. Note: This function is named as this to avoid overload conflict with CheckAndAlloc(const std::vector<TShape> &aux_shapes), since TShape tmp = some_shape is equivalent to TShape tmp = {some_shape}. More...

void	CheckAndAlloc (const std::vector< TShape > &aux_shapes) const

void	CheckAndAllocData (const TShape &storage_shape) const

void	CheckAndAllocAuxData (size_t i, const TShape &aux_shape) const

Static Public Member Functions
static void	Save (dmlc::Stream *fo, const std::vector< NDArray > &data, const std::vector< std::string > &names)
	Save list of ndarray into the Stream.x. More...

static void	Load (dmlc::Stream fi, std::vector< NDArray > data, std::vector< std::string > *keys)
	Load list of ndarray into from the stream. More...

Friends
class	Imperative

Detailed Description

ndarray interface

Constructor & Destructor Documentation

mxnet::NDArray::NDArray ( )

inline

default constructor

mxnet::NDArray::NDArray	(	const TShape &	shape,
		Context	ctx,
		bool	delay_alloc = `false`,
		int	dtype = `mshadow::default_type_flag`
	)

inline

constructs a new dynamic NDArray

Parameters

shape	the shape of array
ctx	context of NDArray
delay_alloc	whether delay the allocation
dtype	data type of this ndarray

mxnet::NDArray::NDArray	(	const NDArrayStorageType	stype,
		const TShape &	shape,
		Context	ctx,
		bool	delay_alloc = `true`,
		int	dtype = `mshadow::default_type_flag`,
		std::vector< int >	aux_types = `{}`,
		std::vector< TShape >	aux_shapes = `{}`,
		TShape	storage_shape = `TShape(mshadow::Shape1(0))`
	)

inline

constructor for NDArray with storage type

mxnet::NDArray::NDArray	(	const TBlob &	data,
		int	dev_id
	)

inline

constructing a static NDArray that shares data with TBlob Use with caution: allocate ONLY ONE NDArray for each TBlob, make sure the memory region is available through out the life of NDArray

Parameters

data	the memory content of static data
dev_id	the device id this tensor sits at

mxnet::NDArray::NDArray	(	const NDArrayStorageType	stype,
		const TShape &	shape,
		const TBlob &	data,
		const std::vector< TBlob > &	aux_data,
		int	dev_id
	)

inline

constructing a static NDArray of non-default storage that shares data with TBlob Use with caution: allocate ONLY ONE NDArray for each TBlob, make sure the memory region is available through out the life of NDArray

Parameters

stype	the storage type of NDArray
shape	the shape of NDArray
data	the memory content of static data
aux_data	the memory content of static aux data
dev_id	the device id this tensor sits at

Member Function Documentation

NDArray mxnet::NDArray::AsArray	(	const TShape &	shape,
		int	dtype
	)		const

inline

Create a NDArray that shares memory with current one The new array must have smaller memory size than the current array.

Parameters

shape	new shape
dtype	The data type.

Returns: NDArray in new shape and type.

NDArray mxnet::NDArray::At ( index_t idx ) const

Index a NDArray.

Parameters

idx the index

Returns: idx-th sub array NDArray

NDArray mxnet::NDArray::AtWithRecord ( index_t idx )

Index a NDArray.

Parameters

idx the index

Returns: idx-th sub array NDArray

TBlob mxnet::NDArray::aux_data ( size_t i ) const

inline

Returns: the aux TBlob

NDArray mxnet::NDArray::aux_ndarray ( size_t i ) const

Generate a deep copy of aux_data(i) returned as a default storage type NDArray.

const TShape& mxnet::NDArray::aux_shape ( size_t index ) const

inline

get the shape of aux_data(index)

Parameters

index the index of the aux data

Returns: the shape of aux data at given index

const std::vector<TShape>& mxnet::NDArray::aux_shapes ( ) const

inline

int mxnet::NDArray::aux_type ( size_t i ) const

inline

const std::vector<int>& mxnet::NDArray::aux_types ( ) const

inline

returns the dtypes of all aux data

void mxnet::NDArray::CheckAndAlloc ( ) const

inline

Allocate the space if it is delayed allocated. This is an internal function used by system that normal user should not use.

void mxnet::NDArray::CheckAndAlloc ( const std::vector< TShape > & aux_shapes ) const

inline

void mxnet::NDArray::CheckAndAllocAuxData	(	size_t	i,
		const TShape &	aux_shape
	)		const

inline

void mxnet::NDArray::CheckAndAllocData ( const TShape & storage_shape ) const

inline

NDArray mxnet::NDArray::Copy ( Context ctx ) const

return a new copy this NDArray

Parameters

ctx	the new context of this NDArray

Returns: the new copy

Context mxnet::NDArray::ctx ( ) const

inline

Returns: the context of NDArray, this function is only valid when the NDArray is not empty

const TBlob& mxnet::NDArray::data ( ) const

inline

Returns: the data TBlob

NDArray mxnet::NDArray::data_ndarray ( ) const

Generate a deep copy of data() returned as a default storage type NDArray.

NDArray mxnet::NDArray::Detach ( ) const

inline

Return a copy of this NDArray without autograd history.

int mxnet::NDArray::dtype ( ) const

inline

Returns: the data type of NDArray, this function is only valid when the NDArray is not empty

bool mxnet::NDArray::fresh_out_grad ( ) const

Returns: updated grad state in entry_

nnvm::Symbol mxnet::NDArray::get_autograd_symbol ( ) const

NDArray mxnet::NDArray::grad ( ) const

Returns: the gradient ndarray.

bool mxnet::NDArray::is_none ( ) const

inline

Returns: whether this ndarray is not initialized

bool mxnet::NDArray::LegacyLoad	(	dmlc::Stream *	strm,
		const uint32_t	magic
	)

load ndarrays before supporting sparse ndarrays

Parameters

strm	the output stream
magic	the magic number used for version control

bool mxnet::NDArray::Load ( dmlc::Stream * strm )

load the content from binary stream

Parameters

strm	the output stream

Returns: whether the load is successful

static void mxnet::NDArray::Load	(	dmlc::Stream *	fi,
		std::vector< NDArray > *	data,
		std::vector< std::string > *	keys
	)

static

Load list of ndarray into from the stream.

Parameters

fi	The stream of the input file.
data	the NDArrays to be loaded
keys	the name of the NDArray, if saved in the file.

NDArray& mxnet::NDArray::operator*= ( const NDArray & src )

elementwise multiplication to current ndarray this mutate the current NDArray

Parameters

src	the data to subtract

Returns: reference of self

NDArray& mxnet::NDArray::operator*= ( const real_t & src )

elementwise multiplication to current ndarray this mutate the current NDArray

Parameters

src	the data to subtract

Returns: reference of self

NDArray& mxnet::NDArray::operator+= ( const NDArray & src )

elementwise add to current space this mutate the current NDArray

Parameters

src	the data to add

Returns: reference of self

NDArray& mxnet::NDArray::operator+= ( const real_t & src )

elementwise add to current space this mutate the current NDArray

Parameters

src	the data to add

Returns: reference of self

NDArray& mxnet::NDArray::operator-= ( const NDArray & src )

elementwise subtract from current ndarray this mutate the current NDArray

Parameters

src	the data to subtract

Returns: reference of self

NDArray& mxnet::NDArray::operator-= ( const real_t & src )

elementwise subtract from current ndarray this mutate the current NDArray

Parameters

src	the data to subtract

Returns: reference of self

NDArray& mxnet::NDArray::operator/= ( const NDArray & src )

elementwise division from current ndarray this mutate the current NDArray

Parameters

src	the data to subtract

Returns: reference of self

NDArray& mxnet::NDArray::operator/= ( const real_t & src )

elementwise division from current ndarray this mutate the current NDArray

Parameters

src	the data to subtract

Returns: reference of self

NDArray& mxnet::NDArray::operator= ( real_t scalar )

set all the elements in ndarray to be scalar

Parameters

scalar the scalar to set

Returns: reference of self

NDArray mxnet::NDArray::Reshape ( const TShape & shape ) const

Get an reshaped NDArray.

Parameters

shape new shape

Returns: NDArray in new shape

void mxnet::NDArray::ReshapeAndAlloc ( const TShape & shape )

inline

Allocate the space if the allocation has been delayed or the requested size is bigger than the available one. This function can only be called by ndarray of default storage type and effectively changes the ndarray's shape_. Note: This function is named as this to avoid overload conflict with CheckAndAlloc(const std::vector<TShape> &aux_shapes), since TShape tmp = some_shape is equivalent to TShape tmp = {some_shape}.

NDArray mxnet::NDArray::ReshapeWithRecord ( const TShape & shape )

Get an reshaped NDArray. Supports autograd recording.

Parameters

shape new shape

Returns: NDArray in new shape

void mxnet::NDArray::Save ( dmlc::Stream * strm ) const

save the content into binary stream

Parameters

strm	the output stream

static void mxnet::NDArray::Save	(	dmlc::Stream *	fo,
		const std::vector< NDArray > &	data,
		const std::vector< std::string > &	names
	)

static

Save list of ndarray into the Stream.x.

Parameters

fo	The stream of output.
data	the NDArrays to be saved.
names	the name of the NDArray, optional, can be zero length.

void mxnet::NDArray::set_aux_shape	(	size_t	index,
		const TShape &	shape
	)		const

inline

For a sparse operation on a csr matrix for example, the size of the column index array is an estimated value in the beginning for allocating enough capacity for the final result. After the operation is done, the exact size of the shape is known and need to be reset using this function.

void mxnet::NDArray::set_fresh_out_grad ( bool state ) const

Returns: updated grad state in entry_

const TShape& mxnet::NDArray::shape ( ) const

inline

Returns: the shape of current NDArray.

NDArray mxnet::NDArray::Slice	(	index_t	begin,
		index_t	end
	)		const

Slice a NDArray.

Parameters

begin	begin index in first dim (inclusive)
end	end index in first dim (exclusive)

Returns: sliced NDArray

NDArray mxnet::NDArray::SliceWithRecord	(	index_t	begin,
		index_t	end
	)

Slice a NDArray. Supports recording with autograd.

Parameters

begin	begin index in first dim (inclusive)
end	end index in first dim (exclusive)

Returns: sliced NDArray

bool mxnet::NDArray::storage_initialized ( ) const

inline

const TShape& mxnet::NDArray::storage_shape ( ) const

inline

Returns: the shape of underlying chunk which stores the NDArray data/value. It is only intended for non-default storage. For row-sparse storage, it is the shape of the tensor which stores the non-zero values.

NDArrayStorageType mxnet::NDArray::storage_type ( ) const

inline

void mxnet::NDArray::SyncCopyFromCPU	(	const void *	data,
		size_t	size
	)		const

Do a synchronize copy from a continugous CPU memory region.

This function will call WaitToWrite before the copy is performed. This is useful to copy data from existing memory region that are not wrapped by NDArray(thus dependency not being tracked).

Parameters

data	the data source to copy from.
size	the size of the source array, in sizeof(DType) not raw btyes.

void mxnet::NDArray::SyncCopyFromNDArray	(	const NDArray &	src,
		int	i = `-1`,
		int	j = `-1`
	)

Copy from src.data()/aux_data(i) to this->data()/aux_data(j)

void mxnet::NDArray::SyncCopyToCPU	(	void *	data,
		size_t	size
	)		const

Do a synchronize copy to a continugous CPU memory region.

This function will call WaitToRead before the copy is performed. This is useful to copy data from existing memory region that are not wrapped by NDArray(thus dependency not being tracked).

Parameters

data	the data source to copyinto.
size	the memory size we want to copy into, in sizeof(DType) not raw btyes.

Engine::VarHandle mxnet::NDArray::var ( ) const

inline

Returns: the associated variable of the ndarray.

void mxnet::NDArray::WaitToRead ( ) const

inline

Block until all the pending write operations with respect to current NDArray are finished, and read can be performed.

void mxnet::NDArray::WaitToWrite ( ) const

inline

Block until all the pending read/write operations with respect to current NDArray are finished, and write can be performed.

Push an empty mutable function to flush all preceding reads to the variable.

Friends And Related Function Documentation

friend class Imperative

friend

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

include/mxnet/ndarray.h

Public Member Functions

Static Public Member Functions

Friends

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Friends And Related Function Documentation