from
Teaching Numerical Gradients and Hessians
by Brendan Wood
Simple, well-commented Matlab code to demonstrate how to take numerical derivatives and Hessians.
|
| fun(X, NFV) |
% fun.m
%
% [y, NFV] = fun(X, NFV)
%
% Test objective function for use with the numerical gradient and Hessian
% functions, and the test script sym_hess.m.
%
% Inputs:
% -> X: Point at which to evaluate the function.
% -> NFV: Accumulator to keep track of number of function
% evaluations.
%
% Outputs:
% -> y: Value of function at X.
% -> NFV: Accumulator incremented by the number of function
% evaluations which have taken place.
%
% Created by: Brendan C. Wood
% Created on: February 14, 2011
%
% Copyright (c) 2011, Brendan C. Wood <b.wood@unb.ca>
%
function [y, NFV] = fun(X, NFV)
y = 2*X(1)^5 -3*X(1)*X(3) + 4*X(2)^2 - 9*X(2)*X(3)^2 + 6*X(1)*X(2)*X(3);
NFV = NFV + 1;
|
|
Contact us
