MultiClassDecisionStumpMultiThreshold< DataType > Struct Template Reference

#include <MultiClassDecisionStump.h>

Public Member Functions
	MultiClassDecisionStumpMultiThreshold (int n_classes)
	initialize the vector memory
void	evaluate_feature (int *response, DataType value) const
	convert the feature value to {-1,+1} using internal threshold

Static Public Member Functions
static std::string	signature ()

Public Attributes
std::vector< std::pair < DataType, int > >	responses

Detailed Description

template<class DataType>
struct MultiClassDecisionStumpMultiThreshold< DataType >

A simple decision stump for multiclass problem using only one threshold for each class NOTE: really hard to implement

Member Data Documentation

template<class DataType>

std::vector<std::pair<DataType, int> > MultiClassDecisionStumpMultiThreshold< DataType >::responses

positive or negative response to the class and associated threshold NOTE: if feature extractor implements "parity" this part can be simplified

---

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

• include/Classifier/MultiClassDecisionStump.h