d = eigs(A)
d = eigs(A, k)
d = eigs(A, k, sigma)
% sigma:
% 'largestabs' (default)
% 'smallestabs'
% 'largestreal'
% 'smallestreal'
% 'bothendsreal'
% 'largestimag'
% 'smallestimag'
% 'bothendsimag'
% scalar
% (No equivalent)
% (No equivalent)
% (No equivalent)
d = eigs(A, k, sigma, Name, Value)
d = eigs(A, k, sigma, opts)
% opts.issym = 1
% opts.tol
% opts.maxit (= 300)
% opts.p
% opts.v0
% opts.disp
% opts.fail
% opts.spdB
% opts.cholB
% opts.permB
% (No equivalent)
d = eigs(A, B, ..)
d = eigs(Afun, n, ..)
[V, D] = eigs(..)
[V, D, flag] = eigs(..)
|
d = eigs(A)
d = eigs(A, [], k)
d = eigs(A, [], k, sigma)
// sigma:
// 'LM' : Largest Magnitudes (default)
// 'SM' : Smallest Magnitudes
// 'LR' : Largest Real parts
// 'SM' : Smallest Real parts
// (No equivalent)
// 'LI' : Largest Imaginary parts (real non-symmetric or complex problems)
// 'SI' : Smallest Imaginary parts (real non-symmetric or complex problems)
// (No equivalent)
// scalar
// 'LA' : Largest Algebraic eigenvalues (real symmetric problems).
// 'SA' : Smallest Algebraic eigenvalues (real symmetric problems).
// 'BE' : half from each end of the spectrum (real symmetric problems).
// No equivalent. Sets and uses the opts structure instead
d = eigs(A, [], k, sigma, opts)
// opts.issym (= %t) : used with Afun
// opts.tol
// opts.maxiter (= 300)
// opts.ncv : Number of Lanczos basis vectors
// opts.resid : starting residual vector
// (No equivalent)
// (No equivalent)
// (No equivalent)
// opts.cholB
// (No equivalent)
// opts.isreal (= %t) : used with Afun
d = eigs(A, B, ..)
d = eigs(Afun, n, ..)
[D, V] = eigs(..) // D still as first output
// No equivalent
|