Exemplo n.º 1
0
        /*
         *  var functionUnderStudy = new Func<double, double>(x => 0.2 * Math.Pow(x, 3) + 0.1 * Math.Pow(x, 2) - 8 * x);
         *  var fitFunction = new Func<double, double>(x => -(0.2 * Math.Pow(x, 3) + 0.1 * Math.Pow(x, 2) - 8 * x));
         *
         *  var functionUnderStudy = new Func<double, double>(x => x - Math.Pow(x, 2) + Math.Pow(x, 3));
         *  var fitFunction = functionUnderStudy;
         *
         *  var functionUnderStudy = new Func<double, double>(Math.Sin);
         *  var fitFunction = functionUnderStudy;
         */
        public static void Main(string[] args)
        {
            var iterations         = 1000;
            var functionUnderStudy = new Func <double[], double>(x => 0.2 * Math.Pow(x[0], 3) + 0.1 * Math.Pow(x[0], 2) - 8 * x[0]);
            var fitFunction        = new Func <double[], double>(x => Math.Sin(x[0]));
            var cd = new ChromosomeDefinition(10);
            var fc = new Function(functionUnderStudy, fitFunction, -7, 7);
            var rd = new ResearchDefinitions(cd, fc, 100, 0.1, 0.5);
            var listOfMediumAbsFitness = new List <double>();

            var startPop = Fitness.FitPop(
                Chromosome.NewRandomPopulation(
                    rd,
                    rd.Population));

            listOfMediumAbsFitness.Add(startPop.Sum(x => x.Fitness) / startPop.Count);

            var preselectedPop = new RouletteWheel().DrawChromosomes(startPop);

            var nextGenPop     = Fitness.FitPop(PostSelection.CreateNewPopulation(preselectedPop, rd.CrossChance, rd.MutationChance));
            var bestChromosome = new BestChromosome()
            {
                bestChromosome = nextGenPop.Max(x => x),
                generation     = 1
            };

            listOfMediumAbsFitness.Add(nextGenPop.Sum(x => x.Fitness) / nextGenPop.Count);

            for (int i = 1; i < iterations; i++)
            {
                nextGenPop = Fitness.FitPop(PostSelection.CreateNewPopulation(nextGenPop, rd.CrossChance, rd.MutationChance));
                if (bestChromosome.bestChromosome.AbsFitness < nextGenPop.Max(x => x.AbsFitness))
                {
                    bestChromosome.bestChromosome = nextGenPop.FirstOrDefault(x => x.AbsFitness == nextGenPop.Max(y => y.AbsFitness));
                    bestChromosome.generation     = i + 1;
                }
                listOfMediumAbsFitness.Add(nextGenPop.Sum(x => x.Fitness) / nextGenPop.Count);
            }

            listOfMediumAbsFitness.ForEach(Console.WriteLine);
        }
Exemplo n.º 2
0
        private void UpdateLoop()
        {
            //int currentTick = Environment.TickCount;
            //int stopAtTick = currentTick + CurrentTimeLimit;

            bool useTimeLimit      = CurrentTimeLimit > 0;
            bool useIterationLimit = CurrentIterationLimit > 0;

            while (Continue)
            {
                if (NeedsUpdate)
                {
                    Sim.RemoveActions();
                    Sim.AddActions(true, BestChromosome.Values.Where(y => y > 0).Select(x => CraftingAction.CraftingActions[x]));
                    NeedsUpdate = false;

                    if (CopyBestRotationToPopulations)
                    {
                        for (int i = 0; i < Populations.Length; i++)
                        {
                            Populations[i].PendingBest = BestChromosome.Clone();
                        }
                    }
                    CraftingSim sim = Sim.Clone(true);
                    Utils.AddRotationFromSim(sim);
                    FoundBetterRotation(sim);
                }

                if ((useIterationLimit && (Iterations >= CurrentIterationLimit)))
                {
                    Continue = false;
                }
            }

            Stopped();
        }
Exemplo n.º 3
0
 public static void WriteBestChromosome(BestChromosome chromosome)
 {
     Console.WriteLine($"Best chromosome from generation {chromosome.generation}: ");
     WriteChromosome(chromosome.bestChromosome);
 }