CSIM: linear_classification Class Reference

linear_classification Class Reference

#include <linear_classification.h>

Inheritance diagram for linear_classification:

List of all members.

Detailed Description

Implementation of a linear classification.

Public Member Functions

linear_classification (unsigned int in_rows=1, unsigned int classes=2)
~linear_classification (void)
int apply (const double *S, double *X)
void reset ()
int importRepresentation (const double *rep, int rep_length)
double * exportRepresentation (int *rep_length)
string getFormatDescription ()

Public Attributes

int nClasses
Number of Classes.
bool addBias
Add bias (yes / no).

Private Member Functions

double weighted_sum (const double *S, const double *w)

Private Attributes

double ** regression_coefficients
Vector of linear-classification coefficients.

Friends

class csimClassInfoDB

Constructor & Destructor Documentation

linear_classification::linear_classification ( unsigned int in_rows = 1,

unsigned int classes = 2

)

Constructs a new linear classification algorithm.
Parameters:

in_rows Number of rows in each input vector.

classes Number of classes for classification task.

linear_classification::~linear_classification ( void )

Frees the memory.

Member Function Documentation

int linear_classification::apply ( const double * S,

double * X

) [virtual]

Applies the currently learned function to the filtered and preprocessed input vector.
Parameters:

S State of the liquid (= filtered and preprocessed response of the neural microcircuit).

X Target pointer where to save the result.

Returns:
-1 if an error occured, 1 for success.

Implements Algorithm.

double * linear_classification::exportRepresentation ( int * rep_length ) [virtual]

Exports the representation of this algorithm for use in external objects.
Parameters:

rep_length Length of the representation vector.

Returns:
A list of parameters that represent the algorithm.

Warning:
Do not forget to free the memory reserved for the representation!

Implements Algorithm.

string linear_classification::getFormatDescription ( ) [virtual]

Returns a textual description of the representation format for import/export - Representation.
Implements Algorithm.

int linear_classification::importRepresentation ( const double * rep,

int rep_length

) [virtual]

Imports the data from an externally (e.g. Matlab) trained algorithm.
Parameters:

rep Representation of the algorithm as a double vector.

rep_length Length of the representation vector.

Returns:
-1 if an error occured, 1 for success.

Implements Algorithm.

void linear_classification::reset ( ) [virtual]

Resets the information stored within the algorithm.
Reimplemented from Algorithm.

double linear_classification::weighted_sum ( const double * S,

const double * w

) [private]

Calculate a weighted sum of input S and weights w.

last modified 07/10/2006