LightGBM::FeatureGroup Class Reference

Using to store data and providing some operations on one feature group. More...

#include <feature_group.h>

Public Member Functions
	FeatureGroup (int num_feature, std::vector< std::unique_ptr< BinMapper > > &bin_mappers, data_size_t num_data, double sparse_threshold, bool is_enable_sparse)
	Constructor.
	FeatureGroup (int num_feature, std::vector< std::unique_ptr< BinMapper > > &bin_mappers, data_size_t num_data, bool is_sparse)
	FeatureGroup (const void *memory, data_size_t num_all_data, const std::vector< data_size_t > &local_used_indices)
	Constructor from memory.
	~FeatureGroup ()
	Destructor.
void	PushData (int tid, int sub_feature_idx, data_size_t line_idx, double value)
	Push one record, will auto convert to bin and push to bin data.
void	CopySubset (const FeatureGroup *full_feature, const data_size_t *used_indices, data_size_t num_used_indices)
BinIterator *	SubFeatureIterator (int sub_feature)
BinIterator *	FeatureGroupIterator ()
	Returns a BinIterator that can access the entire feature group's raw data. The RawGet() function of the iterator should be called for best efficiency.
data_size_t	Split (int sub_feature, const uint32_t *threshold, int num_threshold, bool default_left, data_size_t *data_indices, data_size_t num_data, data_size_t *lte_indices, data_size_t *gt_indices) const
double	BinToValue (int sub_feature_idx, uint32_t bin) const
	From bin to feature value.
void	SaveBinaryToFile (const VirtualFileWriter *writer) const
	Save binary data to file.
size_t	SizesInByte () const
	Get sizes in byte of this object.
FeatureGroup &	operator= (const FeatureGroup &)=delete
	Disable copy.
	FeatureGroup (const FeatureGroup &)=delete
	Disable copy.

Data Fields
friend	Dataset
friend	DatasetLoader

Detailed Description

Using to store data and providing some operations on one feature group.

Constructor & Destructor Documentation

FeatureGroup() [1/2]

LightGBM::FeatureGroup::FeatureGroup ( int  num_feature, std::vector< std::unique_ptr< BinMapper > > &  bin_mappers, data_size_t  num_data, double  sparse_threshold, bool  is_enable_sparse  )

inline

Constructor.

Parameters

num_feature	number of features of this group
bin_mappers	Bin mapper for features
num_data	Total number of data
is_enable_sparse	True if enable sparse feature
sparse_threshold	Threshold for treating a feature as a sparse feature

FeatureGroup() [2/2]

LightGBM::FeatureGroup::FeatureGroup ( const void *  memory, data_size_t  num_all_data, const std::vector< data_size_t > &  local_used_indices  )

inline

Constructor from memory.

Parameters

memory	Pointer of memory
num_all_data	Number of global data
local_used_indices	Local used indices, empty means using all data

Member Function Documentation

BinToValue()

double LightGBM::FeatureGroup::BinToValue ( int  sub_feature_idx, uint32_t  bin  ) const

inline

From bin to feature value.

Parameters

bin

Returns

FeatureGroup value of this bin

FeatureGroupIterator()

BinIterator * LightGBM::FeatureGroup::FeatureGroupIterator ( )

inline

Returns a BinIterator that can access the entire feature group's raw data. The RawGet() function of the iterator should be called for best efficiency.

Returns

A pointer to the BinIterator object

PushData()

void LightGBM::FeatureGroup::PushData ( int  tid, int  sub_feature_idx, data_size_t  line_idx, double  value  )

inline

Push one record, will auto convert to bin and push to bin data.

Parameters

tid	Thread id
idx	Index of record
value	feature value of record

SaveBinaryToFile()

void LightGBM::FeatureGroup::SaveBinaryToFile ( const VirtualFileWriter *  writer ) const

inline

Save binary data to file.

Parameters

file	File want to write

---

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

• External/LightGBM_2.2.3/LightGBM-2.2.3/include/LightGBM/feature_group.h