Home | History | Annotate | Download | only in ceres 
      1 // Ceres Solver - A fast non-linear least squares minimizer
      2 // Copyright 2010, 2011, 2012 Google Inc. All rights reserved.
      3 // http://code.google.com/p/ceres-solver/
      4 //
      5 // Redistribution and use in source and binary forms, with or without
      6 // modification, are permitted provided that the following conditions are met:
      7 //
      8 // * Redistributions of source code must retain the above copyright notice,
      9 //   this list of conditions and the following disclaimer.
     10 // * Redistributions in binary form must reproduce the above copyright notice,
     11 //   this list of conditions and the following disclaimer in the documentation
     12 //   and/or other materials provided with the distribution.
     13 // * Neither the name of Google Inc. nor the names of its contributors may be
     14 //   used to endorse or promote products derived from this software without
     15 //   specific prior written permission.
     16 //
     17 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
     18 // AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
     19 // IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
     20 // ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
     21 // LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
     22 // CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
     23 // SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
     24 // INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
     25 // CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
     26 // ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
     27 // POSSIBILITY OF SUCH DAMAGE.
     28 //
     29 // Author: sameeragarwal (at) google.com (Sameer Agarwal)
     30 
     31 #include <cstddef>
     32 #include "ceres/block_sparse_matrix.h"
     33 #include "ceres/block_structure.h"
     34 #include "ceres/casts.h"
     35 #include "ceres/internal/scoped_ptr.h"
     36 #include "ceres/linear_least_squares_problems.h"
     37 #include "ceres/linear_solver.h"
     38 #include "ceres/schur_complement_solver.h"
     39 #include "ceres/triplet_sparse_matrix.h"
     40 #include "ceres/types.h"
     41 #include "glog/logging.h"
     42 #include "gtest/gtest.h"
     43 
     44 namespace ceres {
     45 namespace internal {
     46 
     47 class SchurComplementSolverTest : public ::testing::Test {
     48  protected:
     49   void SetUpFromProblemId(int problem_id) {
     50     scoped_ptr<LinearLeastSquaresProblem> problem(
     51         CreateLinearLeastSquaresProblemFromId(problem_id));
     52 
     53     CHECK_NOTNULL(problem.get());
     54     A.reset(down_cast<BlockSparseMatrix*>(problem->A.release()));
     55     b.reset(problem->b.release());
     56     D.reset(problem->D.release());
     57 
     58     num_cols = A->num_cols();
     59     num_rows = A->num_rows();
     60     num_eliminate_blocks = problem->num_eliminate_blocks;
     61 
     62     x.reset(new double[num_cols]);
     63     sol.reset(new double[num_cols]);
     64     sol_d.reset(new double[num_cols]);
     65 
     66     LinearSolver::Options options;
     67     options.type = DENSE_QR;
     68 
     69     scoped_ptr<LinearSolver> qr(LinearSolver::Create(options));
     70 
     71     TripletSparseMatrix triplet_A(A->num_rows(),
     72                                   A->num_cols(),
     73                                   A->num_nonzeros());
     74     A->ToTripletSparseMatrix(&triplet_A);
     75 
     76     // Gold standard solutions using dense QR factorization.
     77     DenseSparseMatrix dense_A(triplet_A);
     78     LinearSolver::Summary summary1 =
     79         qr->Solve(&dense_A,
     80                   b.get(),
     81                   LinearSolver::PerSolveOptions(),
     82                   sol.get());
     83 
     84     // Gold standard solution with appended diagonal.
     85     LinearSolver::PerSolveOptions per_solve_options;
     86     per_solve_options.D = D.get();
     87     LinearSolver::Summary summary2 =
     88         qr->Solve(&dense_A,
     89                   b.get(),
     90                   per_solve_options,
     91                   sol_d.get());
     92   }
     93 
     94   void ComputeAndCompareSolutions(
     95       int problem_id,
     96       bool regularization,
     97       ceres::LinearSolverType linear_solver_type,
     98       ceres::SparseLinearAlgebraLibraryType sparse_linear_algebra_library) {
     99     SetUpFromProblemId(problem_id);
    100     LinearSolver::Options options;
    101     options.elimination_groups.push_back(num_eliminate_blocks);
    102     options.elimination_groups.push_back(
    103         A->block_structure()->cols.size() - num_eliminate_blocks);
    104     options.type = linear_solver_type;
    105     options.sparse_linear_algebra_library = sparse_linear_algebra_library;
    106 
    107     scoped_ptr<LinearSolver> solver(LinearSolver::Create(options));
    108 
    109     LinearSolver::PerSolveOptions per_solve_options;
    110     LinearSolver::Summary summary;
    111     if (regularization) {
    112       per_solve_options.D = D.get();
    113     }
    114 
    115     summary = solver->Solve(A.get(), b.get(), per_solve_options, x.get());
    116 
    117     if (regularization) {
    118       for (int i = 0; i < num_cols; ++i) {
    119         ASSERT_NEAR(sol_d.get()[i], x[i], 1e-10);
    120       }
    121     } else {
    122       for (int i = 0; i < num_cols; ++i) {
    123         ASSERT_NEAR(sol.get()[i], x[i], 1e-10);
    124       }
    125     }
    126   }
    127 
    128   int num_rows;
    129   int num_cols;
    130   int num_eliminate_blocks;
    131 
    132   scoped_ptr<BlockSparseMatrix> A;
    133   scoped_array<double> b;
    134   scoped_array<double> x;
    135   scoped_array<double> D;
    136   scoped_array<double> sol;
    137   scoped_array<double> sol_d;
    138 };
    139 
    140 #ifndef CERES_NO_SUITESPARSE
    141 TEST_F(SchurComplementSolverTest, SparseSchurWithSuiteSparse) {
    142   ComputeAndCompareSolutions(2, false, SPARSE_SCHUR, SUITE_SPARSE);
    143   ComputeAndCompareSolutions(3, false, SPARSE_SCHUR, SUITE_SPARSE);
    144   ComputeAndCompareSolutions(2, true, SPARSE_SCHUR, SUITE_SPARSE);
    145   ComputeAndCompareSolutions(3, true, SPARSE_SCHUR, SUITE_SPARSE);
    146 }
    147 #endif  // CERES_NO_SUITESPARSE
    148 
    149 #ifndef CERES_NO_CXSPARSE
    150 TEST_F(SchurComplementSolverTest, SparseSchurWithCXSparse) {
    151   ComputeAndCompareSolutions(2, false, SPARSE_SCHUR, CX_SPARSE);
    152   ComputeAndCompareSolutions(3, false, SPARSE_SCHUR, CX_SPARSE);
    153   ComputeAndCompareSolutions(2, true, SPARSE_SCHUR, CX_SPARSE);
    154   ComputeAndCompareSolutions(3, true, SPARSE_SCHUR, CX_SPARSE);
    155 }
    156 #endif  // CERES_NO_CXSPARSE
    157 
    158 TEST_F(SchurComplementSolverTest, DenseSchur) {
    159   // The sparse linear algebra library type is ignored for
    160   // DENSE_SCHUR.
    161   ComputeAndCompareSolutions(2, false, DENSE_SCHUR, SUITE_SPARSE);
    162   ComputeAndCompareSolutions(3, false, DENSE_SCHUR, SUITE_SPARSE);
    163   ComputeAndCompareSolutions(2, true, DENSE_SCHUR, SUITE_SPARSE);
    164   ComputeAndCompareSolutions(3, true, DENSE_SCHUR, SUITE_SPARSE);
    165 }
    166 
    167 }  // namespace internal
    168 }  // namespace ceres
    169