/*M/////////////////////////////////////////////////////////////////////////////////////// // // IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING. // // By downloading, copying, installing or using the software you agree to this license. // If you do not agree to this license, do not download, install, // copy or use the software. // // // License Agreement // For Open Source Computer Vision Library // // Copyright (C) 2013, OpenCV Foundation, all rights reserved. // Third party copyrights are property of their respective owners. // // Redistribution and use in source and binary forms, with or without modification, // are permitted provided that the following conditions are met: // // * Redistribution's of source code must retain the above copyright notice, // this list of conditions and the following disclaimer. // // * Redistribution's in binary form must reproduce the above copyright notice, // this list of conditions and the following disclaimer in the documentation // and/or other materials provided with the distribution. // // * The name of the copyright holders may not be used to endorse or promote products // derived from this software without specific prior written permission. // // This software is provided by the copyright holders and contributors "as is" and // any express or implied warranties, including, but not limited to, the implied // warranties of merchantability and fitness for a particular purpose are disclaimed. // In no event shall the OpenCV Foundation or contributors be liable for any direct, // indirect, incidental, special, exemplary, or consequential damages // (including, but not limited to, procurement of substitute goods or services; // loss of use, data, or profits; or business interruption) however caused // and on any theory of liability, whether in contract, strict liability, // or tort (including negligence or otherwise) arising in any way out of // the use of this software, even if advised of the possibility of such damage. // //M*/ #include "test_precomp.hpp" namespace opencv_test { namespace { TEST(Core_LPSolver, regression_basic){ cv::Mat A,B,z,etalon_z; #if 1 //cormen's example #1 A=(cv::Mat_(3,1)<<3,1,2); B=(cv::Mat_(3,4)<<1,1,3,30,2,2,5,24,4,1,2,36); std::cout<<"here A goes\n"<(3,1)<<8,4,0); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12); #endif #if 1 //cormen's example #2 A=(cv::Mat_(1,2)<<18,12.5); B=(cv::Mat_(3,3)<<1,1,20,1,0,20,0,1,16); std::cout<<"here A goes\n"<(2,1)<<20,0); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12); #endif #if 1 //cormen's example #3 A=(cv::Mat_(1,2)<<5,-3); B=(cv::Mat_(2,3)<<1,-1,1,2,1,2); std::cout<<"here A goes\n"<(2,1)<<1,0); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12); #endif } TEST(Core_LPSolver, regression_init_unfeasible){ cv::Mat A,B,z,etalon_z; #if 1 //cormen's example #4 - unfeasible A=(cv::Mat_(1,3)<<-1,-1,-1); B=(cv::Mat_(2,4)<<-2,-7.5,-3,-10000,-20,-5,-10,-30000); std::cout<<"here A goes\n"<(3,1)<<1250,1000,0); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12); #endif } TEST(DISABLED_Core_LPSolver, regression_absolutely_unfeasible){ cv::Mat A,B,z,etalon_z; #if 1 //trivial absolutely unfeasible example A=(cv::Mat_(1,1)<<1); B=(cv::Mat_(2,2)<<1,-1); std::cout<<"here A goes\n"<(2,1)<<1,1); B=(cv::Mat_(1,3)<<1,1,1); std::cout<<"here A goes\n"<(4,1)<<10,-57,-9,-24); B=(cv::Mat_(3,5)<<0.5,-5.5,-2.5,9,0,0.5,-1.5,-0.5,1,0,1,0,0,0,1); std::cout<<"here A goes\n"<(3,1)<<3,1,2); Mat B=(cv::Mat_(3,4)<<1,1,3,30,2,2,5,24,4,1,2,36); Mat1f z_float; cv::solveLP(A, B, z_float); Mat1d z_double; cv::solveLP(A, B, z_double); Mat1i z_int; cv::solveLP(A, B, z_int); EXPECT_ANY_THROW(Mat1b z_8u; cv::solveLP(A, B, z_8u)); } // Degenerate problem: the first ratio test is a three-way tie. // The optimum is unique: z = (16, 15, 0, 24) / 37, objective 189 / 37. TEST(Core_LPSolver, issue_12343) { Mat A = (cv::Mat_(4, 1) << 3., 3., 3., 4.); Mat B = (cv::Mat_(4, 5) << 0., 1., 4., 4., 3., 3., 1., 2., 2., 3., 4., 4., 0., 1., 4., 4., 0., 4., 1., 4.); Mat z; ASSERT_EQ(SOLVELP_SINGLE, cv::solveLP(A, B, z)); Mat etalon_z = cv::Mat_({4, 1}, {16. / 37, 15. / 37, 0., 24. / 37}); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12); } // Cormen's example #1 from regression_basic with the right-hand side scaled by 1e-14. // The solution scales with the data: nothing may be treated as zero because it is small. TEST(Core_LPSolver, small_magnitude_data) { const double scale = 1e-14; Mat A = cv::Mat_({3, 1}, {3, 1, 2}); Mat B = cv::Mat_({3, 4}, {1, 1, 3, 30 * scale, 2, 2, 5, 24 * scale, 4, 1, 2, 36 * scale}); Mat z; ASSERT_EQ(SOLVELP_SINGLE, cv::solveLP(A, B, z)); Mat etalon_z = cv::Mat_({3, 1}, {8 * scale, 4 * scale, 0}); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12 * scale); } // A negative right-hand side makes solveLP start with Phase I. The objective is parallel to // the first constraint, so the optimal points are z = (t, 3 - 3t) for 0 <= t <= 0.9. TEST(Core_LPSolver, negative_rhs_restore_cancellation) { Mat A = cv::Mat_({2, 1}, {-3., -1.}); Mat B = cv::Mat_({3, 3}, { -3., -1., -3., -2., -4., -3., 1., 2., 6. }); Mat z; ASSERT_EQ(SOLVELP_MULTI, cv::solveLP(A, B, z)); ASSERT_EQ(2, z.rows); const double t = z.at(0); EXPECT_GE(t, -1e-12); EXPECT_LE(t, 0.9 + 1e-12); EXPECT_NEAR(3 - 3 * t, z.at(1), 1e-12); } // x1 must be 0 and x2 >= 2/3, so the optimum is unique: z = (0, 2/3), objective -8/3. // Phase I ends with its auxiliary variable basic at 0 and a 0 in the first column of its row. TEST(Core_LPSolver, phase1_auxiliary_basic_at_zero) { Mat A = cv::Mat_({2, 1}, {1., -4.}); Mat B = cv::Mat_({2, 3}, { -4., -3., -2., 2., 0., 0. }); Mat z; ASSERT_EQ(SOLVELP_SINGLE, cv::solveLP(A, B, z)); Mat etalon_z = cv::Mat_({2, 1}, {0., 2. / 3}); ASSERT_LT(cvtest::norm(z, etalon_z, cv::NORM_L1), 1e-12); } }} // namespace