[go: up one dir, main page]
Menu ▾ ▴

[r129]: / src / agf_calc_pdf.cc  Maximize  Restore  History

Download this file

140 lines (109 with data), 4.0 kB 

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
//Copyright (C) 2007 Peter Mills.  All rights reserved.



#include <math.h>



#include "kextreme.h"

#include "quicksort.h"

#include "agf_lib.h"



using namespace std;

using namespace libpetey;



//do this the easy way:

namespace libagf {



//estimates the pdf at a test point based on a set of samples:

//

//mat is a set of n vectors with m elements

//vec is the vector we want to classify

//k is the number of nearest neighbours to use in the classification

//minw is the minimum un-normalized total of the weights

//

//returns the pdf

template <class real>

real agf_calc_pdf(real **mat, dim_ta D, nel_ta n, real *vec, real var[2], 

			nel_ta k, real Wc, agf_diag_param *diag_param) {

  real *d2;			//the distances (squared)

  real var_f;			//final value of the filter width (as variance)

  real tw;			//total weight

  real *knearest;		//distances of k nearest neighbours

  real *weight;		//the current value for the weights

  real n1, norm;		//normalisation coeff.

  real pdf;			//final calculated value of pdf



  //first we calculate all the distances:

  d2=new real[n];

  for (nel_ta i=0; i<n; i++) {

    d2[i]=metric2(vec, mat[i], D);

    //printf("%g ", d2[i]);

  }

  //printf("\n");



  //select out the k nearest:

  knearest=new real[k];

  KLEAST_FUNC(d2, n, k, knearest);



  //calculate the weights using the central "engine":

  weight=new real[k];

  diag_param->nd=AGF_CALC_W_FUNC(knearest, k, Wc, var, weight, var_f);

  tw=0;

  for (nel_ta i=0; i<k; i++) tw+=weight[i];



  //use the final filter width to normalize the pdf:

  n1=sqrt(var_f*M_PI*2);

  norm=1;

  for (dim_ta i=0; i<D; i++) norm*=n1;

  

  //norm=pow(var_f*M_PI*2, m/2.);

  //printf("var_f=%g, tw=%g, norm=%g\n", var_f, tw, norm);



  pdf=tw/norm/n;



  //set the diagnostic parameters:

  diag_param->f=weight[k-1]/weight[0];

  diag_param->W=tw;



  delete [] d2;

  delete [] knearest;

  delete [] weight;



  return pdf;



}



//estimates the pdf at a test point based on a set of samples:

//this version uses all of the training data...

//

//mat is a set of n vectors with m elements

//vec is the vector we want to classify

//minw is the minimum un-normalized total of the weights

//

//returns the pdf

template <class real>

real agf_calc_pdf(real **mat, dim_ta D, nel_ta n, real *vec, real var[2], 

			real Wc, agf_diag_param *diag_param) {



  real *d2;			//the distances (squared)

  real var_f;			//final value of the filter width (as variance)

  real tw;			//total weight

  real *weight;		//the current value for the weights

  real n1, norm;		//normalisation coeff.

  real pdf;			//final calculated value of pdf



  //first we calculate all the distances:

  d2=new real[n];

  for (nel_ta i=0; i<n; i++) {

    d2[i]=metric2(vec, mat[i], D);

    //printf("%g ", d2[i]);

  }

  //printf("\n");



  //calculate the weights using the central "engine":

  weight=new real[n];

  diag_param->nd=AGF_CALC_W_FUNC(d2, n, Wc, var, weight, var_f);

  tw=0;

  for (nel_ta i=0; i<n; i++) tw+=weight[i];



  //use the final filter width to normalize the pdf:

  n1=sqrt(var_f*M_PI*2);

  norm=1;

  for (dim_ta i=0; i<D; i++) norm*=n1;

  

  //norm=pow(var_f*M_PI*2, m/2.);

  //printf("var_f=%g, tw=%g, norm=%g\n", var_f, tw, norm);



  pdf=tw/norm/n;



  //set the diagnostic parameters:

  //diag_param->f=0;

  diag_param->W=tw;



  delete [] d2;

  delete [] weight;



  return pdf;



}



template float agf_calc_pdf<float>(float **mat, dim_ta D, nel_ta n, float *vec, 

		float var[2], nel_ta k, float Wc, agf_diag_param *diag_param);

template double agf_calc_pdf<double>(double **mat, dim_ta D, nel_ta n, double *vec, 

		double var[2], nel_ta k, double Wc, agf_diag_param *diag_param);



template float agf_calc_pdf<float>(float **mat, dim_ta D, nel_ta n, float *vec, 

		float var[2], float Wc, agf_diag_param *diag_param);

template double agf_calc_pdf<double>(double **mat, dim_ta D, nel_ta n, 

		double *vec, double var[2], double Wc, agf_diag_param *diag_param);



}