SparseLU

SparseLU.cpp

#include "eigen/Eigen/Dense"
#include "eigen/Eigen/SparseLU"
#include "eigen/Eigen/IterativeLinearSolvers"
#include <iostream>
#include <fstream>

using namespace Eigen;


int main() {
	std::vector<Eigen::Triplet<double>> tripletList;

	std::ifstream ifs("triplets.txt");
	int vertNum;
	ifs >> vertNum;
	while (ifs) {
		int col, row;
		double value;

		ifs >> col >> row >> value;
		if (ifs) {
			tripletList.emplace_back(col, row, value);
		}
	}

	// fill matrix A(n*n)
	SparseMatrix<double, ColMajor> A;
	A.resize(vertNum, vertNum);
	A.setZero();
	A.setFromTriplets(tripletList.begin(), tripletList.end());

	// compress sparse matrix(necessary)
	A.makeCompressed();

	// fill right hand side matrix B(n*3)
	MatrixXd X(vertNum, 2), B(vertNum, 2);

	for (int i = 0; i < vertNum; ++i) {
		B.coeffRef(i, 0) = i / double(vertNum);
		B.coeffRef(i, 1) = 0;
	}


	// Solve sparse linear system with supernodal LU factorization
	SparseLU<SparseMatrix<double, ColMajor>> solver;

	solver.compute(A);
	if (solver.info() != Success) {
		std::cout << "err" << std::endl;
		return 1;
	}

	// solve
	X = solver.solve(B);
	if (solver.info() != Success) {
		std::cout << "err" << std::endl;
		return 1;
	}

	for (int i = 0; i < 10; ++i) {
		std::cout << X.coeffRef(i, 0) << ",";
	}
	std::cout << std::endl;

	// Solve sparse linear system with bi conjugate gradient stabilized algorithm
	BiCGSTAB<SparseMatrix<double, ColMajor>> solver1;
	solver1.setMaxIterations(10000);
	solver1.setTolerance(1.0e-6);

	solver1.compute(A);
	if (solver1.info() != Success) {
		std::cout << "err" << std::endl;
		return 1;
	}
	X = solver1.solve(B);
	if (solver1.info() != Success) {
		std::cout << "err" << std::endl;
		return 1;
	}

	for (int i = 0; i < 10; ++i) {
		std::cout << X.coeffRef(i, 0) << ",";
	}

	return 0;
}