source: trunk/src/PCA.h @ 13

Last change on this file since 13 was 13, checked in by daniel, 19 years ago

* empty log message *

  • Property svn:eol-style set to native
  • Property svn:keywords set to Author Date Id Revision
File size: 2.7 KB
Line 
1#ifndef GENETICS_PCA_ANALYZER_H
2#define GENETICS_PCA_ANALYZER_H
3
4// C++ tools include
5/////////////////////
6#include "vector.h"
7#include "matrix.h"
8#include "SVD.h"
9
10// Standard C++ includes
11////////////////////////
12#include <vector>
13#include <iostream>
14#include <memory>
15#include <cstdlib>
16
17
18namespace thep_cpp_tools
19{
20  /**
21     Class performing PCA using SVD.
22  */
23 
24  class PCA
25  {
26  public:
27    /**
28       This constructor is only to be used in test-class
29    */
30    PCA(); 
31
32    /**
33       Constructor taking the data-matrix as input. No row-centering
34       should have been performed and no products.
35     */
36    explicit PCA( const thep_gsl_api::matrix& );
37
38   
39    /**
40       Will perform PCA according to the following scheme: \n
41       1: Rowcenter A  \n
42       2: SVD(A)  --> USV' \n
43       3: Calculate eigenvalues according to \n
44          \f$ \lambda_{ii} = s_{ii}/N_{rows} \f$ \n
45       4: Sort eigenvectors (from matrix V) according to descending eigenvalues \n
46    */
47    void process();
48
49   
50    /**
51       Performes a simple test on performance. Not optimal!
52       Returns true if ok otherwise false.
53    */
54    bool test();
55
56
57    /**
58       Returns eigenvector \a i
59    */
60    thep_gsl_api::matrix get_eigenvector( const size_t& i ) const
61    {
62      return eigenvectors_.row( i );
63    }
64
65    /**
66       Returns eigenvalues to covariance matrix
67       \f$ C = \frac{1}{N^2}A^TA \f$
68    */
69    double get_eigenvalue( const size_t& i ) const
70    {
71      return eigenvalues_[ i ];
72    }
73
74    /**
75       Returns the explained intensity of component \a K \n
76       \f$I = \frac{ \sum^{K}_{i=1} \lambda_i }{ \sum^{N}_{j=1} \lambda_j }\f$ \n
77       where \f$N\f$ is the dimension
78    */
79    double PCA::get_explained_intensity( const size_t& k );
80
81
82
83    /**
84        This function will project data onto the new coordinate-system
85  where the axes are the calculated eigenvectors. This means that
86  PCA must have been run before this function can be used!
87  Output is presented as coordinates in the N-dimensional room
88  spanned by the eigenvectors.
89    */
90    thep_gsl_api::matrix projection( const thep_gsl_api::matrix& ) const;
91
92   
93  private:
94    thep_gsl_api::matrix A_; 
95    thep_gsl_api::matrix  eigenvectors_;
96    thep_gsl_api::vector  eigenvalues_;
97    thep_gsl_api::vector  explained_intensity_;
98    thep_gsl_api::vector  meanvalues_;
99    bool process_, explained_calc_;
100   
101    /**
102       Private function that will row-center the matrix A,
103       that is, A = A - M, where M is a matrix
104       with the meanvalues of each row
105    */
106    void row_center( thep_gsl_api::matrix& A_center );
107
108    /**
109       Private function that will calculate the explained
110       intensity
111    */
112    void calculate_explained_intensity();
113
114   
115  }; // class PCA 
116 
117}
118
119#endif
120
Note: See TracBrowser for help on using the repository browser.