Dear All,

In some statistical applications, I need to solve L^T x = b (i.e. not A x = L L^T x = b) many times for different bs where L is a sparse Cholesky factor.
Is there any way of using the internal code in solve() to do this? If not, does anyone have any thoughts on how to best implement this operation using Eigen?

Best regards, Tore Kleppe

That's easy:

Code: Select all

SparseLLT<...> llt(A);
x = llt.matrixU().solve(b);
x = llt.permutationPinv() * x;


Thanks! That does the job - Regards, Tore