c Copyright (C) 2008 VZLU Prague, a.s., Czech Republic c c Author: Jaroslav Hajek <highegg@gmail.com> c c This file is part of OctGPR. c c OctGPR is free software; you can redistribute it and/or modify c it under the terms of the GNU General Public License as published by c the Free Software Foundation; either version 2 of the License, or c (at your option) any later version. c c This program is distributed in the hope that it will be useful, c but WITHOUT ANY WARRANTY; without even the implied warranty of c MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the c GNU General Public License for more details. c c You should have received a copy of the GNU General Public License c along with this software; see the file COPYING. If not, see c <http://www.gnu.org/licenses/>. c subroutine nl0gpr(nx,y,nu,nll0,nllinf) c purpose: this function evaluates the negative log likelihood c of a GPR process regressor at boundaries: c nll0 is a common value for theta = 0 and c nllinf is a value for theta = infinity. c nx (in) number of training points c y (in) array of training input values c nu (in) relative white noise. nu = sqrt(var_white/var) c nll0 (out) the value of nllgpr for theta = 0 c nllinf (out) the limit of nllgpr at theta -> infinity. c integer nx double precision y(nx),nu,nll0,nllinf double precision mu,ssq parameter (l2pi = 1.83787706640935d0) integer i mu = 0 c calculate mean do i = 1,nx mu = mu + y(i) end do mu = mu / nx c calculate sigma estimate ssq = 0 do i = 1,nx ssq = ssq + (y(i)-mu)**2 end do c set values ssq = ssq / nx nllinf = 0.5d0 * nx * (log(ssq) + l2pi) nll0 = nllinf + 0.5d0 * log(1 + nx/nu**2) end subroutine

Generated by Doxygen 1.6.0 Back to index