FeaturedProblem

FeaturedProblem(P, F, seed, max_eval, termination_eval)

FeaturedProblem is a subclass of Problem class and defines an optimization problem with a specific feature.

---

Problem and its subclass FeaturedProblem describe the following optimization problem:

\[\begin{split}\min \quad & \mathrm{fun}(x) \\ \text{s.t.} \quad & x_l \le x \le x_u, \\ & A_{\mathrm{ub}} x \le b_{\mathrm{ub}}, \\ & A_{\mathrm{eq}} x = b_{\mathrm{eq}}, \\ & c_{\mathrm{ub}}(x) \le 0, \\ & c_{\mathrm{eq}}(x) = 0, \\ & \text{with initial point } x_0.\end{split}\]

FeaturedProblem should be initialized by the following signature:

FP = FeaturedProblem(P, F, seed, max_eval, termination_eval) creates an instance of FeaturedProblem, where the input P is an instance of Problem, the input F is an instance of Feature, the input seed is a nonnegative integer seed less than 2^32, and the inputs max_eval and termination_eval are positive integers.

The output FP contains the following properties:

• problem: the original optimization problem.

• feature: the feature applied to the optimization problem.

• max_eval: the maximum number of function evaluations.

• seed: the seed for the random number generator.

• fun_hist: the history of the evaluated objective function values.

• cub_hist: the history of the evaluated nonlinear inequality constraints.

• ceq_hist: the history of the evaluated nonlinear equality constraints.

• maxcv_hist: the history of the maximum constraint violation.

• n_eval_fun: the minimum between the number of objective function evaluations and max_eval.

• n_eval_cub: the minimum between the number of nonlinear inequality constraint evaluations and max_eval.

• n_eval_ceq: the minimum between the number of nonlinear equality constraint evaluations and max_eval.

• fun_init: the objective function value at the initial point.

• maxcv_init: the maximum constraint violation at the initial point.

The output FP contains all the methods of Problem, but the methods fun, cub, ceq, and maxcv are modified by the input Feature.

Note the following two points. 1. When the number of function evaluations reaches the input max_eval, the methods fun, cub, and ceq will return the values of the objective function and constraints at the point where the maximum number of function evaluations is reached, respectively. 2. When the number of function evaluations reaches the input termination_eval, the methods fun, cub, and ceq will raise an error to terminate the optimization process.