ThunderSVM
ThunderSVM: An Open-Source SVM Library on GPUs and CPUs
Classes | Public Types | Public Member Functions | List of all members
DataSet Class Reference

Dataset reader. More...

#include <dataset.h>

Classes

struct  node
 

Public Types

typedef vector< vector< DataSet::node > > node2d
 

Public Member Functions

 DataSet (const DataSet::node2d &instances, int n_features, const vector< float_type > &y)
 
void load_from_file (string file_name)
 load dataset from file
 
void load_from_python (float *y, char **x, int len)
 load dataset from python
 
void group_classes (bool classification=true)
 group instances in same class
 
size_t n_instances () const
 
size_t n_features () const
 
size_t n_classes () const
 
const vector< int > & count () const
 the number of instances for each class
 
const vector< int > & start () const
 the start position of instances for each class
 
const vector< int > & label () const
 mapping logical label (0,1,2,3,...) to real label (maybe 2,4,5,6,...)
 
const vector< float_type > & y () const
 label for each instances, the instances are arranged as they are in file
 
const node2d & instances () const
 
const node2d instances (int y_i) const
 instances of class \(y_i\)
 
const node2d instances (int y_i, int y_j) const
 instances of class \(y_i\) and \(y_j\)
 
const vector< int > original_index () const
 mapping instance index (after grouped) to the original index (in file)
 
const vector< int > original_index (int y_i) const
 
const vector< int > original_index (int y_i, int y_j) const
 

Detailed Description

Dataset reader.

Constructor & Destructor Documentation

◆ DataSet()

DataSet::DataSet ( const DataSet::node2d &  instances,
int  n_features,
const vector< float_type > &  y 
)

construct a dataset using given instances

Parameters
instancesgiven instances
n_featuresthe number of features of given instances
ythe label of each instances

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