dmlc::data::CSVParser< IndexType, DType > Class Template Reference

CSVParser, parses a dense csv format. All columns are treated as real dense data. Label will be empty if the label column is not specified. More...

#include <csv_parser.h>

Inheritance diagram for dmlc::data::CSVParser< IndexType, DType >:

Public Member Functions
	CSVParser (InputSplit *source, const std::map< std::string, std::string > &args, int nthread)
Public Member Functions inherited from dmlc::data::TextParserBase< IndexType, DType >
	TextParserBase (InputSplit *source, int nthread)
virtual void	BeforeFirst (void)
	set before first of the item
virtual size_t	BytesRead (void) const
virtual bool	ParseNext (std::vector< RowBlockContainer< IndexType, DType > > *data)
	read in next several blocks of data
Public Member Functions inherited from dmlc::data::ParserImpl< IndexType, DType >
virtual bool	Next (void)
	implement next
virtual const RowBlock< IndexType, DType > &	Value (void) const
	get current data
Public Member Functions inherited from dmlc::Parser< IndexType, DType >
Parser< uint32_t, real_t > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
Parser< uint64_t, real_t > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
Parser< uint32_t, int32_t > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
Parser< uint64_t, int32_t > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
Parser< uint32_t, int64_t > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
Parser< uint64_t, int64_t > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
Public Member Functions inherited from dmlc::DataIter< DType >
virtual	~DataIter (void) DMLC_THROW_EXCEPTION
	destructor

Protected Member Functions
virtual void	ParseBlock (const char *begin, const char *end, RowBlockContainer< IndexType, DType > *out)
	parse data into out
Protected Member Functions inherited from dmlc::data::TextParserBase< IndexType, DType >
bool	FillData (std::vector< RowBlockContainer< IndexType, DType > > *data)
	read in next several blocks of data

Additional Inherited Members
Public Types inherited from dmlc::Parser< IndexType, DType >
typedef Parser< IndexType, DType > *(*	Factory) (const std::string &path, const std::map< std::string, std::string > &args, unsigned part_index, unsigned num_parts)
	Factory type of the parser.
Static Public Member Functions inherited from dmlc::Parser< IndexType, DType >
static Parser< IndexType, DType > *	Create (const char *uri_, unsigned part_index, unsigned num_parts, const char *type)
	create a new instance of parser based on the "type"
Static Protected Member Functions inherited from dmlc::data::TextParserBase< IndexType, DType >
static const char *	BackFindEndLine (const char *bptr, const char *begin)
	start from bptr, go backward and find first endof line
static void	IgnoreUTF8BOM (const char **begin, const char **end)
	Ignore UTF-8 BOM if present.
Protected Attributes inherited from dmlc::data::ParserImpl< IndexType, DType >
IndexType	data_ptr_
	pointer to begin and end of data
IndexType	data_end_
std::vector< RowBlockContainer< IndexType, DType > >	data_
	internal data
RowBlock< IndexType, DType >	block_
	internal row block

Detailed Description

template<typename IndexType, typename DType = real_t>
class dmlc::data::CSVParser< IndexType, DType >

CSVParser, parses a dense csv format. All columns are treated as real dense data. Label will be empty if the label column is not specified.

This should be extended in future to accept arguments of column types.

Member Function Documentation

ParseBlock()

template<typename IndexType , typename DType >

void dmlc::data::CSVParser< IndexType, DType >::ParseBlock ( const char *  begin, const char *  end, RowBlockContainer< IndexType, DType > *  out  )

protectedvirtual

parse data into out

Parameters

begin	beginning of buffer
end	end of buffer

Implements dmlc::data::TextParserBase< IndexType, DType >.

---

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

• External/xgboost/dmlc-core/src/data/csv_parser.h