warp.fem.linalg

Linear algebra utilities for FEM computations.

This module provides kernel-compatible linear algebra functions including matrix decompositions (QR factorization, tensor decompositions), eigenvalue solvers, matrix inversion, and array operations commonly needed in finite element computations.

API

array_axpy	Compute y = alpha*x + beta*y.
generalized_inner	Compute a generalized inner product.
generalized_outer	Compute a generalized outer product.
householder_make_hessenberg	Transform a square matrix to Hessenberg form (single lower diagonal) using Householder reflections.
householder_qr_decomposition	QR decomposition of a square matrix using Householder reflections
inverse_qr	Compute the inverse of a square matrix using QR factorization.
skew_part	Skew part of a 3x3 tensor as the corresponding rotation vector.
solve_triangular	Solve for R x = b where R is an upper triangular matrix.
spherical_part	Spherical part of a square tensor.
symmetric_eigenvalues_qr	Compute the eigenvalues and eigen vectors of a square symmetric matrix A using the QR algorithm
symmetric_part	Symmetric part of a square tensor.
tridiagonal_symmetric_eigenvalues_qr	Compute the eigenvalues and eigen vectors of a symmetric tridiagonal matrix using the Symmetric tridiagonal QR algorithm with implicit Wilkinson shift