Group: | Optimizer |
See also: | bfgs |
Function: | nelmin | |
Description: | nelmin searchs for a minimum of a function. In each iteration step the function is evaluated at a simplex consisting of p+1 points. The simplex contracts until the variance of the evaluated function values is less than eps (or the maximal number of iterations is reached). |
Usage: | x = nelmin (x0, f, maxiter {,eps , step}) | |
Input: | ||
x0 | p x n matrix with n starting vectors | |
f | text, name of the procedure where the function is defined | |
maxiter | integer, maximal number of iterations | |
eps | scalar | |
step | scalar, lenght of initial simplex | |
Output: | ||
x.minimum | p x n matrix with the n minima | |
x.iter | number of iterations | |
x.converged | 1 if the algorithm has converged with every starting vector and 0 if it has not |
maxiter should be greater then the number of dimensions p.
proc(y) = f(x) y = sum(x^2) endp x0 = #(1,1,1)~#(1,2,3) nelmin(x0, "f", 100, 1.0e-6)
Contents of nelmin.minimum [1,] 0.017404 -0.02019 [2,] -0.0070216 0.022459 [3,] 0.016622 0.0042336 Contents of nelmin.iter [1,] 53 Contents of nelmin.converged [1,] 1