source: pacpussensors/trunk/Vislab/lib3dv-1.2.0/lib3dv/eigen/test/eigen2/eigen2_array.cpp

Last change on this file was 136, checked in by ldecherf, 8 years ago

Doc

File size: 5.3 KB
Line 
1// This file is part of Eigen, a lightweight C++ template library
2// for linear algebra. Eigen itself is part of the KDE project.
3//
4// Copyright (C) 2008 Gael Guennebaud <g.gael@free.fr>
5//
6// This Source Code Form is subject to the terms of the Mozilla
7// Public License v. 2.0. If a copy of the MPL was not distributed
8// with this file, You can obtain one at http://mozilla.org/MPL/2.0/.
9
10#include "main.h"
11#include <Eigen/Array>
12
13template<typename MatrixType> void array(const MatrixType& m)
14{
15 /* this test covers the following files:
16 Array.cpp
17 */
18
19 typedef typename MatrixType::Scalar Scalar;
20 typedef typename NumTraits<Scalar>::Real RealScalar;
21 typedef Matrix<Scalar, MatrixType::RowsAtCompileTime, 1> VectorType;
22
23 int rows = m.rows();
24 int cols = m.cols();
25
26 MatrixType m1 = MatrixType::Random(rows, cols),
27 m2 = MatrixType::Random(rows, cols),
28 m3(rows, cols);
29
30 Scalar s1 = ei_random<Scalar>(),
31 s2 = ei_random<Scalar>();
32
33 // scalar addition
34 VERIFY_IS_APPROX(m1.cwise() + s1, s1 + m1.cwise());
35 VERIFY_IS_APPROX(m1.cwise() + s1, MatrixType::Constant(rows,cols,s1) + m1);
36 VERIFY_IS_APPROX((m1*Scalar(2)).cwise() - s2, (m1+m1) - MatrixType::Constant(rows,cols,s2) );
37 m3 = m1;
38 m3.cwise() += s2;
39 VERIFY_IS_APPROX(m3, m1.cwise() + s2);
40 m3 = m1;
41 m3.cwise() -= s1;
42 VERIFY_IS_APPROX(m3, m1.cwise() - s1);
43
44 // reductions
45 VERIFY_IS_APPROX(m1.colwise().sum().sum(), m1.sum());
46 VERIFY_IS_APPROX(m1.rowwise().sum().sum(), m1.sum());
47 if (!ei_isApprox(m1.sum(), (m1+m2).sum()))
48 VERIFY_IS_NOT_APPROX(((m1+m2).rowwise().sum()).sum(), m1.sum());
49 VERIFY_IS_APPROX(m1.colwise().sum(), m1.colwise().redux(internal::scalar_sum_op<Scalar>()));
50}
51
52template<typename MatrixType> void comparisons(const MatrixType& m)
53{
54 typedef typename MatrixType::Scalar Scalar;
55 typedef typename NumTraits<Scalar>::Real RealScalar;
56 typedef Matrix<Scalar, MatrixType::RowsAtCompileTime, 1> VectorType;
57
58 int rows = m.rows();
59 int cols = m.cols();
60
61 int r = ei_random<int>(0, rows-1),
62 c = ei_random<int>(0, cols-1);
63
64 MatrixType m1 = MatrixType::Random(rows, cols),
65 m2 = MatrixType::Random(rows, cols),
66 m3(rows, cols);
67
68 VERIFY(((m1.cwise() + Scalar(1)).cwise() > m1).all());
69 VERIFY(((m1.cwise() - Scalar(1)).cwise() < m1).all());
70 if (rows*cols>1)
71 {
72 m3 = m1;
73 m3(r,c) += 1;
74 VERIFY(! (m1.cwise() < m3).all() );
75 VERIFY(! (m1.cwise() > m3).all() );
76 }
77
78 // comparisons to scalar
79 VERIFY( (m1.cwise() != (m1(r,c)+1) ).any() );
80 VERIFY( (m1.cwise() > (m1(r,c)-1) ).any() );
81 VERIFY( (m1.cwise() < (m1(r,c)+1) ).any() );
82 VERIFY( (m1.cwise() == m1(r,c) ).any() );
83
84 // test Select
85 VERIFY_IS_APPROX( (m1.cwise()<m2).select(m1,m2), m1.cwise().min(m2) );
86 VERIFY_IS_APPROX( (m1.cwise()>m2).select(m1,m2), m1.cwise().max(m2) );
87 Scalar mid = (m1.cwise().abs().minCoeff() + m1.cwise().abs().maxCoeff())/Scalar(2);
88 for (int j=0; j<cols; ++j)
89 for (int i=0; i<rows; ++i)
90 m3(i,j) = ei_abs(m1(i,j))<mid ? 0 : m1(i,j);
91 VERIFY_IS_APPROX( (m1.cwise().abs().cwise()<MatrixType::Constant(rows,cols,mid))
92 .select(MatrixType::Zero(rows,cols),m1), m3);
93 // shorter versions:
94 VERIFY_IS_APPROX( (m1.cwise().abs().cwise()<MatrixType::Constant(rows,cols,mid))
95 .select(0,m1), m3);
96 VERIFY_IS_APPROX( (m1.cwise().abs().cwise()>=MatrixType::Constant(rows,cols,mid))
97 .select(m1,0), m3);
98 // even shorter version:
99 VERIFY_IS_APPROX( (m1.cwise().abs().cwise()<mid).select(0,m1), m3);
100
101 // count
102 VERIFY(((m1.cwise().abs().cwise()+1).cwise()>RealScalar(0.1)).count() == rows*cols);
103 VERIFY_IS_APPROX(((m1.cwise().abs().cwise()+1).cwise()>RealScalar(0.1)).colwise().count().template cast<int>(), RowVectorXi::Constant(cols,rows));
104 VERIFY_IS_APPROX(((m1.cwise().abs().cwise()+1).cwise()>RealScalar(0.1)).rowwise().count().template cast<int>(), VectorXi::Constant(rows, cols));
105}
106
107template<typename VectorType> void lpNorm(const VectorType& v)
108{
109 VectorType u = VectorType::Random(v.size());
110
111 VERIFY_IS_APPROX(u.template lpNorm<Infinity>(), u.cwise().abs().maxCoeff());
112 VERIFY_IS_APPROX(u.template lpNorm<1>(), u.cwise().abs().sum());
113 VERIFY_IS_APPROX(u.template lpNorm<2>(), ei_sqrt(u.cwise().abs().cwise().square().sum()));
114 VERIFY_IS_APPROX(ei_pow(u.template lpNorm<5>(), typename VectorType::RealScalar(5)), u.cwise().abs().cwise().pow(5).sum());
115}
116
117void test_eigen2_array()
118{
119 for(int i = 0; i < g_repeat; i++) {
120 CALL_SUBTEST_1( array(Matrix<float, 1, 1>()) );
121 CALL_SUBTEST_2( array(Matrix2f()) );
122 CALL_SUBTEST_3( array(Matrix4d()) );
123 CALL_SUBTEST_4( array(MatrixXcf(3, 3)) );
124 CALL_SUBTEST_5( array(MatrixXf(8, 12)) );
125 CALL_SUBTEST_6( array(MatrixXi(8, 12)) );
126 }
127 for(int i = 0; i < g_repeat; i++) {
128 CALL_SUBTEST_1( comparisons(Matrix<float, 1, 1>()) );
129 CALL_SUBTEST_2( comparisons(Matrix2f()) );
130 CALL_SUBTEST_3( comparisons(Matrix4d()) );
131 CALL_SUBTEST_5( comparisons(MatrixXf(8, 12)) );
132 CALL_SUBTEST_6( comparisons(MatrixXi(8, 12)) );
133 }
134 for(int i = 0; i < g_repeat; i++) {
135 CALL_SUBTEST_1( lpNorm(Matrix<float, 1, 1>()) );
136 CALL_SUBTEST_2( lpNorm(Vector2f()) );
137 CALL_SUBTEST_3( lpNorm(Vector3d()) );
138 CALL_SUBTEST_4( lpNorm(Vector4f()) );
139 CALL_SUBTEST_5( lpNorm(VectorXf(16)) );
140 CALL_SUBTEST_7( lpNorm(VectorXcd(10)) );
141 }
142}
Note: See TracBrowser for help on using the repository browser.