Suppose we have a m by n matrix A with rank m and a set K⊆{1..n} such that the开发者_运维问答 columns of A indexed by K are linearly independent. Now we want to extend K and find a set L so that k⊆L and columns indexed by L are linearly independent too.
One way of doing it would be to start adding column indexes to K and test if the new set are linearly independent or not by using Gaussian Elimination for example. But is there a better way so that I would not need to test for every index added.
Thank You
Only because no one officially answered yet... can't you just use QR?
Q, R = qr(A)
QR factorization finds orthonormal matrix Q and upper triangular matrix R such that A = QR. It implements the Gram-Schmidt algorithm for finding an orthonormal basis for A.
The columns of Q are orthonormal, hence linearly independent. And the first n columns of Q span the column space of A. So Q should give you what you want.
精彩评论