- float *eigval(float *matrix)
- return vector of eigenvalues
(in ascending order) of symmetric matrix
- float *eigval2(float *matrix)
- return vector of eigenvalues
of general matrix. When the matrix has order
n, the returned vector will be of dimension 2n, containing
first the n real parts (in ascending order),
followed by the corresponding n imaginary parts.
- float *diag(float *matrix)
- as before, but matrix will
become replaced by eigenvectors (the i-th row will be the
eigenvector for the i-th eigenvalue in the returned vector).
- float *autocorr(float *data, int stride, float *center=NULL):
- compute
autocorrelation matrix: considers data as concatenation of subvectors
of length stride. Computes and returns stride x stride dimensional
autocorrelation matrix of these subvectors. If center is non-NULL
and at least of dimension stride, the subvectors will be centered
(but data itself will not become changed)
and the subtracted ( stride-dimensional) average will become
written into vector center.
FILE
/local/homes/rhaschke/nst7/man/../o.linx86//../nstsrc/nst_prog_libs.c