Linear-Algebra

The Utilities.cu and Utilities.cuh files are mantained at

https://github.com/OrangeOwlSolutions/CUDA_Utilities

and omitted here.

• Null space calculation by the SVD;
• Calculation of the full SVD of a matrix;
• Calculation of the full QR decomposition of a matrix and inversion of a linear system by the QR decomposition approach (see QR decomposition to solve linear systems in CUDA);
• Calculation of the LU decomposition of a matrix (see LU Decomposition in CUDA);
• Use of LU decomposition to invert linear systems (see Solving linear systems AX = B with CUDA);
• Calculation of the determinant of a matrix using Cholesky decomposition and Thrust strided reduction (see Determinant calculation with CUDA);
• Solve general sparse linear system using cuSOLVER (see Solving general sparse linear systems in CUDA);
• Solve tridiagonal linear systems in CUDA using cuSPARSE (see Solving tridiagonal linear systems in CUDA);
• Solve sparse linear systems by LU factorization (see Solving sparse linear systems in CUDA using LU factorization);
• Solve sparse linear systems by LU factorization and graph coloring (see Improving the solution of sparse linear systems);
• Solve sparse positive definite linear systems by Cholesky factorization (see Solving sparse definite positive linear systems in CUDA);