[Scipy-tickets] [SciPy] #1778: fmin_ncg fails at Rosenbrock

SciPy Trac scipy-tickets@scipy....
Fri Nov 23 19:54:25 CST 2012


#1778: fmin_ncg fails at Rosenbrock
------------------------------+---------------------------------------------
 Reporter:  unknown (google)  |       Owner:  somebody   
     Type:  defect            |      Status:  new        
 Priority:  normal            |   Milestone:  Unscheduled
Component:  scipy.optimize    |     Version:  devel      
 Keywords:  fmin_ncg          |  
------------------------------+---------------------------------------------

Comment(by josefpktd):

 Problem is that I don't get any warning that maxiter has been reached and
 search hasn't converged.

 With large maxiter, fmin_ncg converges, but is very slow with these
 starting values

 {{{
 >>> res = optimize.fmin_ncg(optimize.rosen, np.array([-1.2, 1.0]),
 fprime=optimize.rosen_der, fhess=optimize.rosen_hess, epsilon=1e-5,
 avextol=1e-13, full_output=1, maxiter=1000000)
 Optimization terminated successfully.
          Current function value: 0.000000
          Iterations: 563837
          Function evaluations: 1127622
          Gradient evaluations: 563837
          Hessian evaluations: 563837
 >>> res
 (array([ 1.,  1.]), 3.9535997449180672e-18, 1127622, 563837, 563837, 0)
 >>> import scipy
 >>> scipy.__version__
 '0.9.0'
 }}}

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1778#comment:1>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list