Utility classes

These classes are used by the optimizers in PDFO. The bound, linear, and nonlinear constraints are represented by the Bounds, LinearConstraint, and NonlinearConstraint classes, respectively. Similar classes are also available in scipy.optimize, except that the ones below do not have derivative arguments, as PDFO does not use derivatives. You can use either these classes or those provided by SciPy when calling pdfo, bobyqa, lincoa, and cobyla.

Bounds([lb, ub])	Simple bound constraints.
LinearConstraint([a, lb, ub])	Linear inequality and equality constraints.
NonlinearConstraint(fun[, lb, ub])	Nonlinear inequality and equality constraints.
OptimizeResult	Result structure of the DFO algorithms.