Object oriented genetic algorithm code
The beginnings of code to play around with genetic algorithms.
AbstractEcosystem: -houses the Environment and Specimen -contains main functions to operate alg: Cull, Mate, Mutate, Reproduce -helper functions to mess with env and spec
AbstractEnvironment: -only basic helper functions -intended to be almost entirely implementation specific and affect specimen competition -ex: maze to run through, larger scale game engine (see mtgAlg)
AbstractSpecimen: -contains genes to store solution and fitness to measure weight of solution -likely to have implementation spefic functions to compete and calculate fitness and pass on genes
Uniwalker Implementation: -walkers move right and up to get from bottom left to top right of weighted grid -fitness based on sum of weights of points along the path -unisolver is a O(N!) recursive solver for the minimal weighted path
TestGeneticAlg reads in an input file and runs the code (only uniwalker for now)