Closed Breush closed 5 years ago
I'd like to see "determinist" spelled "deterministic". The former word, "one who determines" is seldom used and is jarring to the ear in our context.
On Fri, Mar 8, 2019 at 5:50 AM A. Breust notifications@github.com wrote:
With this minimal example, we get x = [ 0 0 0 ] as a solution, which is wrong.
include <linbox/solutions/solve.h>
using namespace LinBox; int main(void) { using Ring = Givaro::ZRing
; using Matrix = DenseMatrix ; using Vector = DenseVector ; Ring ZZ; Matrix A(ZZ, 1, 3); Vector x(ZZ, A.coldim()); Vector b(ZZ, A.rowdim()); Ring::Element d; A.setEntry(0, 1, 1); A.setEntry(0, 2, 2); ZZ.assign(b[0], 1); // Calling Dixon solve(x, d, A, b, Method::Dixon(Method::Dixon::SolutionType::DETERMINIST)); std::cout << "[ 0 1 2 ] x = [ 1 ] => xNum/xDen = [ " << x[0] << " " << x[1] << " " << x[2] << " ] /" << d << std::endl; return 0;
}
First investigation shows that FFPACK::LQUPtoInverseOfFullRankMinor within RationalSolver<...>::monolithicSolve might be the culprit.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/linbox-team/linbox/issues/189, or mute the thread https://github.com/notifications/unsubscribe-auth/ADk6I51RgOPw3MFxxtoTVfQkKg50CSMHks5vUkBlgaJpZM4blHpV .
@bdsaunders All right, I changed "Determinist" to "Deterministic" in my WIP branch #169.
Solved by fflas-ffpack PR#230
With this minimal example, we get
x = [ 0 0 0 ]
as a solution, which is wrong.First investigation shows that
FFPACK::LQUPtoInverseOfFullRankMinor
withinRationalSolver<...>::monolithicSolve
might be the culprit.