Hi all, I'm interested in a global optimisation problem with a non-differentiable objective function with about 25 parameters. I've set the options as:
opt.algorithm=NLOPT_G_MLSL_LDS; opt.lower_bounds=lb; opt.upper_bounds=ub; opt.verbose=1; opt.population=10; opt.maxeval=1000; opt.local_optimizer.algorithm=NLOPT_LN_NELDERMEAD; opt.local_optimizer.xtol_rel=1e-3; opt.local_optimizer.ftol_rel=1e-3; opt.local_optimizer.ftol_abs=1e-3; opt.min_objective=obj_func; In this setup, I'm wondering if the opt.population=10 means that the total number of sample points from which the global optimization will start is 10, or if every iteration of the Nelder-Mead method will sample 10 new points? If it's the second, can we figure out, after the optimization is done, how many points were sampled? Finally, I'd appreciate recommendations on the local optimization algorithm I should be using for this problem. Thanks a ton! Best, Nikhil _______________________________________________ NLopt-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/nlopt-discuss
