Multiple sparse linear algebra backends.
1. Added support for CXSparse - SparseNormalCholesky and
SchurComplementSolver support SuiteSparse and CXSparse now.
I am not sure I will add suport for visibility based
preconditioning using CXSparse. Its not a high priority.
2. New enum SparseLinearAlgebraLibraryType which allows the user
to indicate which sparse linear algebra library should be used.
3. Updated tests for SolverImpl and system_test.
4. Build system changes to automatically detect CXSparse and
link to it by default -- just like SuiteSparse.
5. Minor bug fixes dealing in the cmake files and VBP.
6. Changed the order of the system test.
7. Deduped the unsymmetric linear solver test.
Change-Id: I33252a103c87b722ecb7ed7b5f0ae7fd91249244
diff --git a/include/ceres/solver.h b/include/ceres/solver.h
index be81768..701f0b2 100644
--- a/include/ceres/solver.h
+++ b/include/ceres/solver.h
@@ -67,11 +67,18 @@
function_tolerance = 1e-6;
gradient_tolerance = 1e-10;
parameter_tolerance = 1e-8;
-#ifndef CERES_NO_SUITESPARSE
- linear_solver_type = SPARSE_NORMAL_CHOLESKY;
-#else
+
+#if defined(CERES_NO_SUITESPARSE) && defined(CERES_NO_CXSPARSE)
linear_solver_type = DENSE_QR;
-#endif // CERES_NO_SUITESPARSE
+#else
+ linear_solver_type = SPARSE_NORMAL_CHOLESKY;
+#endif
+
+ sparse_linear_algebra_library = SUITE_SPARSE;
+#if defined(CERES_NO_SUITESPARSE) && !defined(CERES_NO_CXSPARSE)
+ sparse_linear_algebra_library = CX_SPARSE;
+#endif
+
preconditioner_type = JACOBI;
num_linear_solver_threads = 1;
num_eliminate_blocks = 0;
@@ -154,6 +161,12 @@
// Type of preconditioner to use with the iterative linear solvers.
PreconditionerType preconditioner_type;
+ // Ceres supports using multiple sparse linear algebra libraries
+ // for sparse matrix ordering and factorizations. Currently,
+ // SUITE_SPARSE and CX_SPARSE are the valid choices, depending on
+ // whether they are linked into Ceres at build time.
+ SparseLinearAlgebraLibraryType sparse_linear_algebra_library;
+
// Number of threads used by Ceres to solve the Newton
// step. Currently only the SPARSE_SCHUR solver is capable of
// using this setting.
