Public Member Functions

CMulticlassAccuracy Class Reference

Detailed Description

The class MulticlassAccuracy used to compute accuracy of multiclass classification.

Formally, for multiclass labels $L,R, |L|=|R|$ multiclass accuracy is estimated as

\[ \frac{\sum_{i=1}^{|L|} [L_i=R_i]}{|L|} \]

Definition at line 34 of file MulticlassAccuracy.h.

Inheritance diagram for CMulticlassAccuracy:
Inheritance graph

List of all members.

Public Member Functions

 CMulticlassAccuracy ()
virtual ~CMulticlassAccuracy ()
virtual float64_t evaluate (CLabels *predicted, CLabels *ground_truth)
EEvaluationDirection get_evaluation_direction ()
virtual const char * get_name () const

Constructor & Destructor Documentation


Definition at line 38 of file MulticlassAccuracy.h.

virtual ~CMulticlassAccuracy (  )  [virtual]


Definition at line 41 of file MulticlassAccuracy.h.

Member Function Documentation

float64_t evaluate ( CLabels predicted,
CLabels ground_truth 
) [virtual]

evaluate accuracy

predicted labels for evaluating
ground_truth labels assumed to be correct

Implements CEvaluation.

Definition at line 17 of file MulticlassAccuracy.cpp.

EEvaluationDirection get_evaluation_direction (  )  [virtual]
whether criterium has to be maximized or minimized

Implements CEvaluation.

Definition at line 50 of file MulticlassAccuracy.h.

virtual const char* get_name (  )  const [virtual]

get name

Implements CSGObject.

Definition at line 56 of file MulticlassAccuracy.h.

The documentation for this class was generated from the following files:
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines

SHOGUN Machine Learning Toolbox - Documentation