geqrf overload (1) - 2023.2 English

Vitis Libraries

Release Date
2023.2 English
#include "MatrixDecomposition/geqrf.hpp"
template <
    typename T,
    int NRMAX,
    int NCMAX,
    int NCU
int geqrf (
    int m,
    int n,
    T* A,
    int lda,
    T* tau

This function computes QR decomposition of matrix \(A\)

\[\begin{equation*} {A = Q R}\end{equation*}\]

where \(A\) is a dense matrix of size \(m \times n\) , \(Q\) is a \(m \times n\) matrix with orthonormal columns, and \(R\) is an upper triangular matrix.

The maximum matrix size supported in FPGA is templated by NRMAX and NCMAX.


T data type (support float and double)
NRMAX maximum number of rows of input matrix
NCMAX maximum number of columns of input matrix
NCU number of computation unit, NCU = n will be n times faster than NCU = 1, and cost n times resource.
m number of rows of matrix A
n number of cols of matrix A
A input matrix of size \(m \times lda\) , and overwritten by the output triangular R matrix and min(m,n) elementary reflectors
lda leading dimension of matrix A
tau scalar factors for elementary reflectors