Struct QRFactorization 

pub struct QRFactorization<S: Scalar + Entity> { /* private fields */ }

QR factorization of a matrix.

Caches the decomposition computed in new() and reuses it in every solve() / solve_least_squares() call, avoiding redundant recomputation.

Implementations

impl<S: Scalar + SimpleEntity + Conjugate<Canonical = S> + ComplexField> QRFactorization<S>

pub fn new(matrix: &DenseMatrix<S>) -> Result<Self, LinalgError>

Compute QR factorization of a matrix.

Requires m >= n (the matrix must have at least as many rows as columns). The decomposition is cached and reused for subsequent solve() / solve_least_squares() calls.

pub fn nrows(&self) -> usize

Number of rows of the original matrix.

pub fn ncols(&self) -> usize

Number of columns of the original matrix.

pub fn solve(&self, b: &[S]) -> Result<Vec<S>, LinalgError>

Solve Ax = b using the cached QR factorization (square systems only).

pub fn solve_least_squares(&self, b: &[S]) -> Result<Vec<S>, LinalgError>

Solve the least-squares problem min ||Ax - b||_2.

Works for both square and overdetermined systems (m >= n). Uses the cached QR factorization.