Files
ceres-solver/internal/ceres/sparse_normal_cholesky_solver.cc
T

111 lines
4.1 KiB
C++
Raw Normal View History

2012-04-30 23:09:08 -07:00
// Ceres Solver - A fast non-linear least squares minimizer
2017-06-13 00:12:00 -07:00
// Copyright 2017 Google Inc. All rights reserved.
// http://ceres-solver.org/
2012-04-30 23:09:08 -07:00
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// * Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
// * Redistributions 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.
// * Neither the name of Google Inc. nor the names of its contributors may 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 COPYRIGHT OWNER 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.
//
// Author: sameeragarwal@google.com (Sameer Agarwal)
#include "ceres/sparse_normal_cholesky_solver.h"
#include <algorithm>
#include <cstring>
#include <ctime>
2018-03-30 16:16:59 -07:00
#include <memory>
2012-05-29 00:27:57 -07:00
2017-06-13 00:12:00 -07:00
#include "ceres/block_sparse_matrix.h"
#include "ceres/inner_product_computer.h"
2013-02-01 12:22:53 -08:00
#include "ceres/internal/eigen.h"
2012-04-30 23:09:08 -07:00
#include "ceres/linear_solver.h"
2017-04-12 12:48:44 -07:00
#include "ceres/sparse_cholesky.h"
2012-04-30 23:09:08 -07:00
#include "ceres/triplet_sparse_matrix.h"
#include "ceres/types.h"
2013-02-01 12:22:53 -08:00
#include "ceres/wall_time.h"
2012-04-30 23:09:08 -07:00
namespace ceres {
namespace internal {
2017-04-25 09:32:08 -07:00
2012-04-30 23:09:08 -07:00
SparseNormalCholeskySolver::SparseNormalCholeskySolver(
const LinearSolver::Options& options)
2017-04-12 12:48:44 -07:00
: options_(options) {
sparse_cholesky_.reset(
SparseCholesky::Create(options_.sparse_linear_algebra_library_type,
options_.use_postordering ? AMD : NATURAL));
2012-04-30 23:09:08 -07:00
}
2017-04-12 12:48:44 -07:00
SparseNormalCholeskySolver::~SparseNormalCholeskySolver() {}
2014-04-26 07:42:23 +01:00
2012-04-30 23:09:08 -07:00
LinearSolver::Summary SparseNormalCholeskySolver::SolveImpl(
2017-06-13 00:12:00 -07:00
BlockSparseMatrix* A,
2012-04-30 23:09:08 -07:00
const double* b,
const LinearSolver::PerSolveOptions& per_solve_options,
2017-04-25 09:32:08 -07:00
double* x) {
2017-04-12 12:48:44 -07:00
EventLogger event_logger("SparseNormalCholeskySolver::Solve");
LinearSolver::Summary summary;
summary.num_iterations = 1;
summary.termination_type = LINEAR_SOLVER_SUCCESS;
summary.message = "Success.";
const int num_cols = A->num_cols();
VectorRef(x, num_cols).setZero();
A->LeftMultiply(b, x);
2017-04-12 12:48:44 -07:00
event_logger.AddEvent("Compute RHS");
if (per_solve_options.D != NULL) {
// Temporarily append a diagonal block to the A matrix, but undo
// it before returning the matrix to the user.
2018-03-30 16:16:59 -07:00
std::unique_ptr<BlockSparseMatrix> regularizer;
2017-06-13 00:12:00 -07:00
regularizer.reset(BlockSparseMatrix::CreateDiagonalMatrix(
per_solve_options.D, A->block_structure()->cols));
event_logger.AddEvent("Diagonal");
A->AppendRows(*regularizer);
2017-06-13 00:12:00 -07:00
event_logger.AddEvent("Append");
}
2017-04-12 12:48:44 -07:00
event_logger.AddEvent("Append Rows");
2017-06-13 00:12:00 -07:00
if (inner_product_computer_.get() == NULL) {
inner_product_computer_.reset(
InnerProductComputer::Create(*A, sparse_cholesky_->StorageType()));
event_logger.AddEvent("InnerProductComputer::Create");
2017-04-25 09:32:08 -07:00
}
2017-06-13 00:12:00 -07:00
inner_product_computer_->Compute();
event_logger.AddEvent("InnerProductComputer::Compute");
// TODO(sameeragarwal):
2012-05-29 00:27:57 -07:00
if (per_solve_options.D != NULL) {
2017-06-13 00:12:00 -07:00
A->DeleteRowBlocks(A->block_structure()->cols.size());
}
2017-04-12 12:48:44 -07:00
summary.termination_type = sparse_cholesky_->FactorAndSolve(
2017-06-13 00:12:00 -07:00
inner_product_computer_->mutable_result(), x, x, &summary.message);
2017-04-12 12:48:44 -07:00
event_logger.AddEvent("Factor & Solve");
return summary;
2012-05-29 00:27:57 -07:00
}
2017-04-25 09:32:08 -07:00
} // namespace internal
} // namespace ceres