Hello everyone,

I do have a strange problem concerning SparseMatrix Iterators.

I declare a SparseMatrix as:

Eigen::SparseMatrix<double, 1> Matrix = function_which_generates_matrix();

for(int k = 0; k < Matrix.outerSize(); k++)
{
for(Eigen::SparseMatrix<double>::InnerIterator it(Matrix,k); it; ++it)
{
std::cout << it.row() << "\t" << it.col() << std::endl;
}
}

This returns something like:

0 0
1 0
2 0
etc.

While I would have expected:

0 0
0 1
0 2
etc.

or something similar.

The strange thing is, this doesn't change at all when I change
Eigen::SparseMatrix<double, 1> = function_which_generates_matrix();
to Eigen::SparseMatrix<double, 0> = function_which_generates_matrix(); or (and here I am not sure whether automatic conversion between storage schemes works in any way)
Eigen::SparseMatrix<double, 0> = function_which_generates_matrix().transpose();

However Eigen::SparseMatrix<double, 0> = function_which_generates_matrix().transpose(); results in:

560912 0
0 0
-50243 0
etc. so this obviously points to some memory access problems.

Do you have any ideas where these problems might come from?

Look at this line:

for(Eigen::SparseMatrix<double>::InnerIterator it(Matrix,k); it; ++it)

the SparseMatrix type does not match the type of 'Matrix', therefore 'Matrix' is copied into a column major matrix when passed to the iterator.