>> opt04_run --------------------------------------------------------- Running testcase_4: Himmelblau --------------------------------------------------------- Newton method: Gradient is within tolerance, x likely a local optimum. Optimization required 13 function/gradient evaluations Newton method produced (3.5844283e+00,-1.8481265e+00) Value of F(X) = 0.000000 Use the secant method: (Algorithm likely to fail!) Maximum number of iterations reached Optimization required 16 function/gradient evaluations BFGS method produced (-4.1909208e+01,-4.3909253e+01) Value of F(X) = 6426075.061546 --------------------------------------------------------- Running testcase_4 as least squares problem: --------------------------------------------------------- Gradient is within tolerance, x likely a local optimum. Optimization required 7 function/gradient evaluations Gauss-Newton produced (3.0000000e+00, 2.0000000e+00) Norm of RES(X) = 0.000000 >>