bench_svd.cpp
Go to the documentation of this file.
1 // This file is part of Eigen, a lightweight C++ template library
2 // for linear algebra.
3 //
4 // Copyright (C) 2013 Gauthier Brun <brun.gauthier@gmail.com>
5 // Copyright (C) 2013 Nicolas Carre <nicolas.carre@ensimag.fr>
6 // Copyright (C) 2013 Jean Ceccato <jean.ceccato@ensimag.fr>
7 // Copyright (C) 2013 Pierre Zoppitelli <pierre.zoppitelli@ensimag.fr>
8 //
9 // This Source Code Form is subject to the terms of the Mozilla
10 // Public License v. 2.0. If a copy of the MPL was not distributed
11 // with this file, You can obtain one at http://mozilla.org/MPL/2.0/
12 
13 // Bench to compare the efficiency of SVD algorithms
14 
15 #include <iostream>
16 #include <bench/BenchTimer.h>
17 #include <unsupported/Eigen/SVD>
18 
19 
20 using namespace Eigen;
21 using namespace std;
22 
23 // number of computations of each algorithm before the print of the time
24 #ifndef REPEAT
25 #define REPEAT 10
26 #endif
27 
28 // number of tests of the same type
29 #ifndef NUMBER_SAMPLE
30 #define NUMBER_SAMPLE 2
31 #endif
32 
33 template<typename MatrixType>
34 void bench_svd(const MatrixType& a = MatrixType())
35 {
36  MatrixType m = MatrixType::Random(a.rows(), a.cols());
37  BenchTimer timerJacobi;
38  BenchTimer timerBDC;
39  timerJacobi.reset();
40  timerBDC.reset();
41 
42  cout << " Only compute Singular Values" <<endl;
43  for (int k=1; k<=NUMBER_SAMPLE; ++k)
44  {
45  timerBDC.start();
46  for (int i=0; i<REPEAT; ++i)
47  {
48  BDCSVD<MatrixType> bdc_matrix(m);
49  }
50  timerBDC.stop();
51 
52  timerJacobi.start();
53  for (int i=0; i<REPEAT; ++i)
54  {
55  JacobiSVD<MatrixType> jacobi_matrix(m);
56  }
57  timerJacobi.stop();
58 
59 
60  cout << "Sample " << k << " : " << REPEAT << " computations : Jacobi : " << fixed << timerJacobi.value() << "s ";
61  cout << " || " << " BDC : " << timerBDC.value() << "s " <<endl <<endl;
62 
63  if (timerBDC.value() >= timerJacobi.value())
64  cout << "KO : BDC is " << timerJacobi.value() / timerBDC.value() << " times faster than Jacobi" <<endl;
65  else
66  cout << "OK : BDC is " << timerJacobi.value() / timerBDC.value() << " times faster than Jacobi" <<endl;
67 
68  }
69  cout << " =================" <<endl;
70  std::cout<< std::endl;
71  timerJacobi.reset();
72  timerBDC.reset();
73  cout << " Computes rotaion matrix" <<endl;
74  for (int k=1; k<=NUMBER_SAMPLE; ++k)
75  {
76  timerBDC.start();
77  for (int i=0; i<REPEAT; ++i)
78  {
80  }
81  timerBDC.stop();
82 
83  timerJacobi.start();
84  for (int i=0; i<REPEAT; ++i)
85  {
87  }
88  timerJacobi.stop();
89 
90 
91  cout << "Sample " << k << " : " << REPEAT << " computations : Jacobi : " << fixed << timerJacobi.value() << "s ";
92  cout << " || " << " BDC : " << timerBDC.value() << "s " <<endl <<endl;
93 
94  if (timerBDC.value() >= timerJacobi.value())
95  cout << "KO : BDC is " << timerJacobi.value() / timerBDC.value() << " times faster than Jacobi" <<endl;
96  else
97  cout << "OK : BDC is " << timerJacobi.value() / timerBDC.value() << " times faster than Jacobi" <<endl;
98 
99  }
100  std::cout<< std::endl;
101 }
102 
103 
104 
105 int main(int argc, char* argv[])
106 {
107  std::cout<< std::endl;
108 
109  std::cout<<"On a (Dynamic, Dynamic) (6, 6) Matrix" <<std::endl;
110  bench_svd<Matrix<double,Dynamic,Dynamic> >(Matrix<double,Dynamic,Dynamic>(6, 6));
111 
112  std::cout<<"On a (Dynamic, Dynamic) (32, 32) Matrix" <<std::endl;
113  bench_svd<Matrix<double,Dynamic,Dynamic> >(Matrix<double,Dynamic,Dynamic>(32, 32));
114 
115  //std::cout<<"On a (Dynamic, Dynamic) (128, 128) Matrix" <<std::endl;
116  //bench_svd<Matrix<double,Dynamic,Dynamic> >(Matrix<double,Dynamic,Dynamic>(128, 128));
117 
118  std::cout<<"On a (Dynamic, Dynamic) (160, 160) Matrix" <<std::endl;
119  bench_svd<Matrix<double,Dynamic,Dynamic> >(Matrix<double,Dynamic,Dynamic>(160, 160));
120 
121  std::cout<< "--------------------------------------------------------------------"<< std::endl;
122 
123 }
void bench_svd(const MatrixType &a=MatrixType())
Definition: bench_svd.cpp:34
Definition: LDLT.h:16
int main(int argc, char *argv[])
Definition: bench_svd.cpp:105
class Bidiagonal Divide and Conquer SVD
Two-sided Jacobi SVD decomposition of a rectangular matrix.
#define NUMBER_SAMPLE
Definition: bench_svd.cpp:30
The matrix class, also used for vectors and row-vectors.
Definition: Matrix.h:178
#define REPEAT
Definition: bench_svd.cpp:25


hebiros
Author(s): Xavier Artache , Matthew Tesch
autogenerated on Thu Sep 3 2020 04:08:02