Meaning of 'minimization_failures' when using scipy.optimize.basinhopping?

Question

I am trying to use scipy.optimize.basinhopping with a function b_log(x) at point x=10. I do not understand the meaning of "minimization_failures" below. Can anyone explain?

 In [144]: scipy.optimize.basinhopping(b_log,10)
    Out[144]:
                      nfev: 6969
     minimization_failures: 101
                       fun: 420
                         x: array([10])
                   message: ['requested number of basinhopping iterations completed successfully']
                      njev: 1919
                       nit: 100

score 2 · Accepted Answer · edited May 23 '17 at 12:09

From the original source:

# do a local minimization
minres = self.minimizer(x_after_step) 
x_after_quench = minres.x
energy_after_quench = minres.fun
if not minres.success:
    self.res.minimization_failures += 1 
    if self.disp:
        print("warning: basinhopping: local minimization failure")

So, minimization_failures means exactly what its name says: The number of times a minimization could not be done during the monte-carlo-step.

Edit: Have a look at this for some explanation of the method. I think u may need to provide the additional arguments to the function (e.g. T and minimizer_kwargs).

Meaning of 'minimization_failures' when using scipy.optimize.basinhopping?

1 Answers1