Public Member Functions

yadt::conf_matrix Class Reference

Models a confusion matrix. More...

#include <YaDT.h>

List of all members.

Public Member Functions

float cases () const
 Return total weights in the confusion matrix.
conf_matrixclone () const
 Clone method.
 conf_matrix (const conf_matrix &)
 Copy constructor not defined.
double get_elapsed () const
 Return elapsed time taken to build the confusion matrix.
float get_element (size_t actual, size_t predicted) const
 Return a cell of the matrix.
float mis_perc () const
 Return misclassification percentage of the confusion matrix.
const conf_matrixoperator= (const conf_matrix &)
 Assignment constructor not defined.
size_t size () const
 Return degree of the matrix.
void toTEXT (std::ostream &os=std::cout, size_t space=0) const
 Text output of the confusion matrix.
void toXML (std::ostream &os=std::cout) const
 XML output of the confusion matrix.
 ~conf_matrix ()
 Destructor.

Detailed Description

Models a confusion matrix.

A confusion matrix is a matrix n x n of weight_type::type, where n is the number of distinct values of a class (including the "unknown" class value). For a conf_matrix object cm, cm.get_element(act, pred) represents the weighted sum of cases of class with index act predicted as belonging to class of index pred. The "unknown" class value has index 0: cm.get_element(0, x) should be 0 since actual classes should never be "unknown" (this may happen, however, if the actual class of cases in the test set did not appear in the training set); cm.get_element(x, 0) should be 0 since predicted classes should never be "unknown". A confusion matrix can only be obtained as the return value of some computation.

See also:
dtree::predict

Constructor & Destructor Documentation

yadt::conf_matrix::~conf_matrix (  ) 

Destructor.

yadt::conf_matrix::conf_matrix ( const conf_matrix  ) 

Copy constructor not defined.


Member Function Documentation

float yadt::conf_matrix::cases (  )  const

Return total weights in the confusion matrix.

More formally, for a conf_matrix object cm, it is returned sum(true, cm[i, j]).

conf_matrix* yadt::conf_matrix::clone (  )  const

Clone method.

Return a newly allocated conf_matrix that is a clone of the called object.

double yadt::conf_matrix::get_elapsed (  )  const

Return elapsed time taken to build the confusion matrix.

float yadt::conf_matrix::get_element ( size_t  actual,
size_t  predicted 
) const

Return a cell of the matrix.

float yadt::conf_matrix::mis_perc (  )  const

Return misclassification percentage of the confusion matrix.

The misclassification percentage of a conf_matrix object cm is sum(i != j, cm[i, j]).

const conf_matrix& yadt::conf_matrix::operator= ( const conf_matrix  ) 

Assignment constructor not defined.

size_t yadt::conf_matrix::size (  )  const

Return degree of the matrix.

The degree of a matrix n x n, is n.

void yadt::conf_matrix::toTEXT ( std::ostream &  os = std::cout,
size_t  space = 0 
) const

Text output of the confusion matrix.

Parameters:
os output stream.
space indent space.
void yadt::conf_matrix::toXML ( std::ostream &  os = std::cout  )  const

XML output of the confusion matrix.


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