class to construct input split with global shuffling More...

#include <input_split_shuffle.h>

Inheritance diagram for dmlc::InputSplitShuffle:

Collaboration diagram for dmlc::InputSplitShuffle:

Public Member Functions
virtual	~InputSplitShuffle (void)

virtual void	BeforeFirst (void)
	reset the position of InputSplit to beginning More...

virtual void	HintChunkSize (size_t chunk_size)
	hint the inputsplit how large the chunk size it should return when implementing NextChunk this is a hint so may not be enforced, but InputSplit will try adjust its internal buffer size to the hinted value More...

virtual size_t	GetTotalSize (void)
	get the total size of the InputSplit More...

virtual bool	NextRecord (Blob *out_rec)
	get the next record, the returning value is valid until next call to NextRecord, NextChunk or NextBatch caller can modify the memory content of out_rec More...

virtual bool	NextChunk (Blob *out_chunk)
	get a chunk of memory that can contain multiple records, the caller needs to parse the content of the resulting chunk, for text file, out_chunk can contain data of multiple lines for recordio, out_chunk can contain multiple records(including headers) More...

virtual void	ResetPartition (unsigned rank, unsigned nsplit)
	reset the Input split to a certain part id, The InputSplit will be pointed to the head of the new specified segment. This feature may not be supported by every implementation of InputSplit. More...

	InputSplitShuffle (const char uri, unsigned part_index, unsigned num_parts, const char type, unsigned num_shuffle_parts, int shuffle_seed)
	constructor More...

Public Member Functions inherited from dmlc::InputSplit
virtual bool	NextBatch (Blob *out_chunk, size_t n_records)
	get a chunk of memory that can contain multiple records, with hint for how many records is needed, the caller needs to parse the content of the resulting chunk, for text file, out_chunk can contain data of multiple lines for recordio, out_chunk can contain multiple records(including headers) More...

virtual	~InputSplit (void) DMLC_THROW_EXCEPTION
	destructor More...

Static Public Member Functions
static InputSplit *	Create (const char uri, unsigned part_index, unsigned num_parts, const char type, unsigned num_shuffle_parts, int shuffle_seed)
	factory function: create input split with chunk shuffling given a uri More...

Static Public Member Functions inherited from dmlc::InputSplit
static InputSplit *	Create (const char uri, unsigned part_index, unsigned num_parts, const char type)
	factory function: create input split given a uri More...

static InputSplit *	Create (const char uri, const char index_uri, unsigned part_index, unsigned num_parts, const char *type, const bool shuffle=false, const int seed=0, const size_t batch_size=256, const bool recurse_directories=false)
	factory function: create input split given a uri for input and index More...

Detailed Description

class to construct input split with global shuffling

Constructor & Destructor Documentation

virtual dmlc::InputSplitShuffle::~InputSplitShuffle ( void )

inlinevirtual

dmlc::InputSplitShuffle::InputSplitShuffle	(	const char *	uri,
		unsigned	part_index,
		unsigned	num_parts,
		const char *	type,
		unsigned	num_shuffle_parts,
		int	shuffle_seed
	)

inline

constructor

Parameters

uri	the uri of the input, can contain hdfs prefix
part_index	the part id of current input
num_parts	total number of splits
type	type of record List of possible types: "text", "recordio" "text": text file, each line is treated as a record input split will split on '\n' or '\r' "recordio": binary recordio file, see recordio.h
num_shuffle_parts	number of shuffle chunks for each split
shuffle_seed	shuffle seed for chunk shuffling

Member Function Documentation

virtual void dmlc::InputSplitShuffle::BeforeFirst ( void )

inlinevirtual

reset the position of InputSplit to beginning

Implements dmlc::InputSplit.

static InputSplit* dmlc::InputSplitShuffle::Create	(	const char *	uri,
		unsigned	part_index,
		unsigned	num_parts,
		const char *	type,
		unsigned	num_shuffle_parts,
		int	shuffle_seed
	)

inlinestatic

factory function: create input split with chunk shuffling given a uri

Parameters

uri	the uri of the input, can contain hdfs prefix
part_index	the part id of current input
num_parts	total number of splits
type	type of record List of possible types: "text", "recordio" "text": text file, each line is treated as a record input split will split on '\n' or '\r' "recordio": binary recordio file, see recordio.h
num_shuffle_parts	number of shuffle chunks for each split
shuffle_seed	shuffle seed for chunk shuffling

Returns: a new input split

See also: InputSplit::Type

virtual size_t dmlc::InputSplitShuffle::GetTotalSize ( void )

inlinevirtual

get the total size of the InputSplit

Implements dmlc::InputSplit.

virtual void dmlc::InputSplitShuffle::HintChunkSize ( size_t chunk_size )

inlinevirtual

hint the inputsplit how large the chunk size it should return when implementing NextChunk this is a hint so may not be enforced, but InputSplit will try adjust its internal buffer size to the hinted value

Parameters

chunk_size the chunk size

Reimplemented from dmlc::InputSplit.

virtual bool dmlc::InputSplitShuffle::NextChunk ( Blob * out_chunk )

inlinevirtual

get a chunk of memory that can contain multiple records, the caller needs to parse the content of the resulting chunk, for text file, out_chunk can contain data of multiple lines for recordio, out_chunk can contain multiple records(including headers)

This function ensures there won't be partial record in the chunk caller can modify the memory content of out_chunk, the memory is valid until next call to NextRecord, NextChunk or NextBatch

Usually NextRecord is sufficient, NextChunk can be used by some multi-threaded parsers to parse the input content

Parameters

out_chunk used to store the result

Returns: true if we can successfully get next record false if we reached end of split

See also: InputSplit::Create for definition of record; RecordIOChunkReader to parse recordio content from out_chunk

Implements dmlc::InputSplit.

virtual bool dmlc::InputSplitShuffle::NextRecord ( Blob * out_rec )

inlinevirtual

get the next record, the returning value is valid until next call to NextRecord, NextChunk or NextBatch caller can modify the memory content of out_rec

For text, out_rec contains a single line For recordio, out_rec contains one record content(with header striped)

Parameters

out_rec used to store the result

Returns: true if we can successfully get next record false if we reached end of split

See also: InputSplit::Create for definition of record

Implements dmlc::InputSplit.

virtual void dmlc::InputSplitShuffle::ResetPartition	(	unsigned	part_index,
		unsigned	num_parts
	)

inlinevirtual

reset the Input split to a certain part id, The InputSplit will be pointed to the head of the new specified segment. This feature may not be supported by every implementation of InputSplit.

Parameters

part_index	The part id of the new input.
num_parts	The total number of parts.

Implements dmlc::InputSplit.

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

/work/mxnet/3rdparty/dmlc-core/include/dmlc/input_split_shuffle.h

Public Member Functions

Static Public Member Functions

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation