Subset of eigenvalues and eigenvectors
Matlab | Scilab |
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 |