Performs singular value decomposition to compute the four fundamental subspaces of a matrix, namely the right and left null spaces and the right and left image spaces.
sv = svd(a)
sv = svd(a, 'econ')
[u, s, v] = svd(a)
[u, s, v] = svd(a, 'econ')
Performs the decomposition in economical sized format.
Singular values of a. sv is a real vector with min(m, n) elements.
m-by-min(m, n) unitary matrix.
Square matrix of order min(m, n) with the singular values on the main diagonal and zeros elsewhere.
n-by-min(m, n) unitary matrix.
Singular value decomposition is a computationally expensive but powerful algorithm for solving a number of problems, including finding least square solutions, finding the 2-norm and 2-norm condition estimate, and determining the rank of a matrix. svd computes unitary matrices u and v, such that the input matrix is equivalent to u*s*conjugate(v').
A = [1, 2, 3, 4; 5, 6, 7, 8; 9, 0, 1, 2; 3, 4, 5, 6]; C = svd(A)
Where This Node Can Run:
Desktop OS: Windows
FPGA: This product does not support FPGA devices