Guys, i ran the simple SVD code from the examples and i get the following error:
I am using Ubuntu 9.1 with g++ compiler and i have pretty much forgotten c++
What simple stuff i am missing. Some other examples did work by the way.

Following is the code:
#include </home/syed/eigen2/Eigen/Core>
#include </home/syed/eigen2/Eigen/Array>
#include </home/syed/eigen2/Eigen/SVD>

MatrixXf A = MatrixXf::Random(20,20);
VectorXf b = VectorXf::Random(20);
VectorXf x;
A.svd().solve(b, &x);
SVD<MatrixXf> svdOfA(A);
svdOfA.solve(b, &x);

This is what i get:
syed@syed-desktop:~/eigen2$ g++ lu.cpp -o lu.exe
lu.cpp:5: error: expected constructor, destructor, or type conversion before ‘=’ token
lu.cpp:6: error: expected constructor, destructor, or type conversion before ‘=’ token
lu.cpp:7: error: ‘VectorXf’ does not name a type
lu.cpp:8: error: expected constructor, destructor, or type conversion before ‘.’ token
lu.cpp:9: error: expected constructor, destructor, or type conversion before ‘<’ token
lu.cpp:10: error: expected constructor, destructor, or type conversion before ‘.’ token

Just put:

Code: Select all

using namespace Eigen;


Looking at the docs, i dont think they need to be fixed, the code doesn't claim to be anything more than an incomplete snippet.

Well i added the code "using namespace Eigen;" and now i am getting the following error. So whats wrong now:

CODE:
#include </home/syed/eigen2/Eigen/Core>
#include </home/syed/eigen2/Eigen/Array>
#include </home/syed/eigen2/Eigen/SVD>
using namespace Eigen;
MatrixXf A = MatrixXf::Random(20,20);
VectorXf b = VectorXf::Random(20);
VectorXf x;
A.svd().solve(b, &x);
SVD<MatrixXf> svdOfA(A);
svdOfA.solve(b, &x);

ERROR:
svd1.cpp:8: error: expected constructor, destructor, or type conversion before ‘.’ token
svd1.cpp:10: error: expected constructor, destructor, or type conversion before ‘.’ token

Indeed you get these erros when you try to compile this code as-is... that's because you need to put the code in a main() function! So just put int main() {........} and that works.

The tutorials, and more generally using Eigen, requires some familiarity with c++, that's true.