fminunc() [coursera] Machine learning - Stanford University - Andrew Ng

Advanced Optimization

fminunc() in Octave

costFunction.m

function [jVal, gradient] = costFunction(theta)
  jVal = (theta(1)-5)^2 + (theta(2)-5)^2;
  gradient = zeros(2,1);
  gradient(1) = 2*(theta(1)-5);
  gradient(2) = 2*(theta(2)-5);
endfunction

prompt

>> options = optimset('GradObj', 'on', 'MaxIter', '100');
>> initialTheta = zeros(2,1);
>> initialTheta
initialTheta =

   0
   0

>> [optTheta, functionVal, exitFlag] = fminunc(@costFunction, initialTheta, options)
optTheta =

   5.0000
   5.0000

functionVal =    1.5777e-30
exitFlag =  1
>> %% exitFlag == 1, means converged
>> %% intialTheta must be an at least 2-dimensional vector