[RBFOpt] Getting more output data

[Giacomo Nannicini]

Fredrik,
you cannot get the parameters from the command line interface.
If you are writing your own Python wrapper and have access to the
RbfoptAlgorithm object after the optimization, then you can retrieve
the parameters, but it will take a bit of work.

Something like this should do it (I did not test this code, I am
writing off the top of my head so you may have to do some fixes, but I
hope you get the idea):

import rbfopt
import rbfopt.rbfopt_utils as ru

# I assume alg is the RbfoptAlgorithm object after optimization, i.e.,
after calling alg.optimize()
settings = RbfoptSettings(rbf=alg.best_global_rbf[0],
rbf_shape_parameter=alg.best_global_rbf[1])
n = alg.n
k = len(alg.node_pos)
matrix = ru.get_rbf_matrix(settings, n, k, alg.node_pos)
rbf_l, rbf_h = ru.get_rbf_coefficients(settings, n, k, Amat, alg.node_val)
# Now rbf_l, rbf_h contain the parameters of the RBF model. You can
evaluate it at a new point with:
new_value = ru.evaluate_rbf(settings, point_at_which_to_evaluate, n,
k, alg.node_pos, rbf_l, rbf_h)
# If you want to evaluate it at many points, you can use this faster version:
new_values = ru.bulk_evaluate_rbf(settings,
points_at_which_to_evaluate, n, k, alg.node_pos, rbf_l, rbf_h)


On Wed, Mar 6, 2019 at 11:36 AM Fredrik Nilsson
<frnilss at student.chalmers.se> wrote:
>
> Hello,
>
> I am using RBFOpt to calibrate a traffic simulation to real world detector data and the relationship between the input parameters and the output is extremely non-linear. Is it possible in any way to get the parameters used in the surrogate model to be output? Either visually or just equations/numbers.
>
> BR,
> Fredrik
> _______________________________________________
> RBFOpt mailing list
> RBFOpt at list.coin-or.org
> https://list.coin-or.org/mailman/listinfo/rbfopt