The whole optimization problem is encoded into a fitness function, which receives a chromosome and returns a number that tells the fitness (or goodness) of the solution. You also specify a range of permissible values (e.g. Those chromosomes consist of an array of genes, that can be bits/ints/floats. In a genetic algorithm, you start with an initial population of chromosomes, which are possible solutions to a given problem. The knowledge can however be applied to other libraries/custom implementations as well, so keep reading ! I am limiting myself to those available in PyGAD, a nice Python library implementing GA. This article is for those like me who haven't done GA ( Genetic Algorithms) in a while, and need a refresher on the most common parent selection, crossover, and mutation algorithms.
