devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Clang-format tests, examples, libraries, benchmarks, etc.

Browse Source
This commit is contained in:
Antonio Sánchez
2023-12-05 21:22:55 +00:00
committed by Rasmus Munk Larsen
parent 3252ecc7a4
commit 46e9cdb7fe
876 changed files with 33453 additions and 37795 deletions
Download Patch File Download Diff File Expand all files Collapse all files

94
test/accelerate_support.cpp
View File

@@ -7,45 +7,43 @@
#include <Eigen/AccelerateSupport>
template<typename MatrixType,typename DenseMat>
int generate_sparse_rectangular_problem(MatrixType& A, DenseMat& dA, int maxRows = 300, int maxCols = 300)
{
template <typename MatrixType, typename DenseMat>
int generate_sparse_rectangular_problem(MatrixType& A, DenseMat& dA, int maxRows = 300, int maxCols = 300) {
typedef typename MatrixType::Scalar Scalar;
int rows = internal::random<int>(1, maxRows);
int cols = internal::random<int>(1, maxCols);
double density = (std::max)(8.0 / (rows * cols), 0.01);
A.resize(rows,cols);
dA.resize(rows,cols);
A.resize(rows, cols);
dA.resize(rows, cols);
initSparse<Scalar>(density, dA, A, ForceNonZeroDiag);
A.makeCompressed();
return rows;
}
template<typename MatrixType,typename DenseMat>
int generate_sparse_square_symmetric_problem(MatrixType& A, DenseMat& dA, int maxSize = 300)
{
template <typename MatrixType, typename DenseMat>
int generate_sparse_square_symmetric_problem(MatrixType& A, DenseMat& dA, int maxSize = 300) {
typedef typename MatrixType::Scalar Scalar;
int rows = internal::random<int>(1, maxSize);
int cols = rows;
double density = (std::max)(8.0 / (rows * cols), 0.01);
A.resize(rows,cols);
dA.resize(rows,cols);
A.resize(rows, cols);
dA.resize(rows, cols);
initSparse<Scalar>(density, dA, A, ForceNonZeroDiag);
dA = dA * dA.transpose();
A = A * A.transpose();
A = A * A.transpose();
A.makeCompressed();
return rows;
}
template<typename Scalar, typename Solver> void test_accelerate_ldlt()
{
typedef SparseMatrix<Scalar> MatrixType;
typedef Matrix<Scalar,Dynamic,1> DenseVector;
template <typename Scalar, typename Solver>
void test_accelerate_ldlt() {
typedef SparseMatrix<Scalar> MatrixType;
typedef Matrix<Scalar, Dynamic, 1> DenseVector;
MatrixType A;
Matrix<Scalar,Dynamic,Dynamic> dA;
Matrix<Scalar, Dynamic, Dynamic> dA;
generate_sparse_square_symmetric_problem(A, dA);
@@ -54,8 +52,7 @@ template<typename Scalar, typename Solver> void test_accelerate_ldlt()
Solver solver;
solver.compute(A);
if (solver.info() != Success)
{
if (solver.info() != Success) {
std::cerr << "sparse LDLT factorization failed\n";
exit(0);
return;
@@ -63,25 +60,24 @@ template<typename Scalar, typename Solver> void test_accelerate_ldlt()
DenseVector x = solver.solve(b);
if (solver.info() != Success)
{
if (solver.info() != Success) {
std::cerr << "sparse LDLT factorization failed\n";
exit(0);
return;
}
//Compare with a dense solver
// Compare with a dense solver
DenseVector refX = dA.ldlt().solve(b);
VERIFY((A * x).isApprox(A * refX, test_precision<Scalar>()));
}
template<typename Scalar, typename Solver> void test_accelerate_llt()
{
typedef SparseMatrix<Scalar> MatrixType;
typedef Matrix<Scalar,Dynamic,1> DenseVector;
template <typename Scalar, typename Solver>
void test_accelerate_llt() {
typedef SparseMatrix<Scalar> MatrixType;
typedef Matrix<Scalar, Dynamic, 1> DenseVector;
MatrixType A;
Matrix<Scalar,Dynamic,Dynamic> dA;
Matrix<Scalar, Dynamic, Dynamic> dA;
generate_sparse_square_symmetric_problem(A, dA);
@@ -90,8 +86,7 @@ template<typename Scalar, typename Solver> void test_accelerate_llt()
Solver solver;
solver.compute(A);
if (solver.info() != Success)
{
if (solver.info() != Success) {
std::cerr << "sparse LLT factorization failed\n";
exit(0);
return;
@@ -99,25 +94,24 @@ template<typename Scalar, typename Solver> void test_accelerate_llt()
DenseVector x = solver.solve(b);
if (solver.info() != Success)
{
if (solver.info() != Success) {
std::cerr << "sparse LLT factorization failed\n";
exit(0);
return;
}
//Compare with a dense solver
// Compare with a dense solver
DenseVector refX = dA.llt().solve(b);
VERIFY((A * x).isApprox(A * refX, test_precision<Scalar>()));
}
template<typename Scalar, typename Solver> void test_accelerate_qr()
{
typedef SparseMatrix<Scalar> MatrixType;
typedef Matrix<Scalar,Dynamic,1> DenseVector;
template <typename Scalar, typename Solver>
void test_accelerate_qr() {
typedef SparseMatrix<Scalar> MatrixType;
typedef Matrix<Scalar, Dynamic, 1> DenseVector;
MatrixType A;
Matrix<Scalar,Dynamic,Dynamic> dA;
Matrix<Scalar, Dynamic, Dynamic> dA;
generate_sparse_rectangular_problem(A, dA);
@@ -126,8 +120,7 @@ template<typename Scalar, typename Solver> void test_accelerate_qr()
Solver solver;
solver.compute(A);
if (solver.info() != Success)
{
if (solver.info() != Success) {
std::cerr << "sparse QR factorization failed\n";
exit(0);
return;
@@ -135,22 +128,20 @@ template<typename Scalar, typename Solver> void test_accelerate_qr()
DenseVector x = solver.solve(b);
if (solver.info() != Success)
{
if (solver.info() != Success) {
std::cerr << "sparse QR factorization failed\n";
exit(0);
return;
}
//Compare with a dense solver
// Compare with a dense solver
DenseVector refX = dA.colPivHouseholderQr().solve(b);
VERIFY((A * x).isApprox(A * refX, test_precision<Scalar>()));
}
template<typename Scalar>
void run_tests()
{
typedef SparseMatrix<Scalar> MatrixType;
template <typename Scalar>
void run_tests() {
typedef SparseMatrix<Scalar> MatrixType;
test_accelerate_ldlt<Scalar, AccelerateLDLT<MatrixType, Lower> >();
test_accelerate_ldlt<Scalar, AccelerateLDLTUnpivoted<MatrixType, Lower> >();
@@ -169,8 +160,7 @@ void run_tests()
test_accelerate_qr<Scalar, AccelerateQR<MatrixType> >();
}
EIGEN_DECLARE_TEST(accelerate_support)
{
EIGEN_DECLARE_TEST(accelerate_support) {
CALL_SUBTEST_1(run_tests<float>());
CALL_SUBTEST_2(run_tests<double>());
}