scikit-opt

Why we included this project

scikit-opt collects a wide set of classic metaheuristic search methods into one Python API: genetic algorithms, particle swarm, simulated annealing, ant colony optimization, differential evolution, and a few more. That makes it a handy starting point for anyone tuning parameters, planning routes, or wrestling with constraint-heavy design problems without coding each search strategy from scratch. Engineers and researchers get the most out of it because swapping optimizers usually means changing the solver class rather than rewriting the objective function, so you can compare several approaches against the same problem quickly. The library also lets you plug in your own selection, crossover, or mutation operators when the defaults are not what you need. For an expensive black-box objective, it is a convenient toolbox for prototyping and for seeing how each method behaves.