Projects

pyevolve

PyevolveDescriptionThis project focus on development of a complete cross-platform (Windows, Linux) framework for Genetic Algorithms in pure python. The project is in active development, and I'm doing great efforts to release a good version and documentation soon. Main features: Many selector ... [More] algorithms like: Rank Selector Uniform Selector Roulette Wheel Selector Tournament Selector Pure python, you can use in any platform which runs python; Easy to use, easy for end-user; Simple to customize Create new representations, genetic operators; You can view the Tutorial - Christian S. Perone News 05/08/2008 Added documentation to the code; Created API Docs. 04/08/2008 I've created a Tutorial. 31/07/2008 Released Pyevolve 0.3 (changes in ChangeLogVersion03)! 29/06/2008 Good news, Pyevolve was figured in the "10 Must-Have Python Packages for Social Scientists" ! 25/01/2008 Released Pyevolve 0.2 (changes in ChangeLogVersion02)! Some code example (SimpleCodeExample)from pyevolve import G1DList from pyevolve import GSimpleGA from pyevolve import Selectors def eval_func(ind): score = 0.0 for x in range(0,len(ind)): if ind[x]==0: score += 0.1 return score # Genome instance genome = G1DList.G1DList(7) genome.setInitParams(rangemin=0, rangemax=6) # The evaluator function (objective function) genome.evaluator.set(eval_func) # Genetic Algorithm Instance ga = GSimpleGA.GSimpleGA(genome) ga.selector.set(Selectors.GRouletteWheel) # Do the evolution ga.evolve() # Best individual print ga.bestIndividual() And the result: - GenomeBase Evaluated: True Score: 0.70 Fitness: 0.70 Slot name: Evaluator (Count: 1) Name: eval_func Doc: No documentation ! Slot name: Initializator (Count: 1) Name: G1DListInitializatorInteger Doc: This is the init function !! Slot name: Mutator (Count: 1) Name: G1DListMutatorSwap Doc: No documentation ! Slot name: Crossover (Count: 1) Name: G1DListCrossoverSinglePoint Doc: No documentation ! - G1DList List size: 7 List: [0, 0, 0, 0, 0, 0, 0] [Less]