public double[] GetFinalScores(Schedule sch, SchedulingFitness fitness)
{
ctx.FillDetails = true;
sch.Details = Enumerable.Range(0, 100).Select(i => new FinalExamDetail()).ToArray();
var results = fitness.CostFunctions.Select(cf => cf(sch)).ToList();
//List<double> results = new List<double>
//{
// fitness.GetStudentDuplicatedScore(sch),
// fitness.GetPresidentNotAvailableScore(sch),
// fitness.GetSecretaryNotAvailableScore(sch),
// fitness.GetExaminerNotAvailableScore(sch),
// fitness.GetMemberNotAvailableScore(sch),
// fitness.GetSupervisorNotAvailableScore(sch),
// fitness.GetPresidentChangeScore(sch),
// fitness.GetSecretaryChangeScore(sch),
// fitness.GetPresidentWorkloadWorstScore(sch),
// fitness.GetPresidentWorkloadWorseScore(sch),
// fitness.GetPresidentWorkloadBadScore(sch),
// fitness.GetSecretaryWorkloadWorstScore(sch),
// fitness.GetSecretaryWorkloadWorseScore(sch),
// fitness.GetSecretaryWorkloadBadScore(sch),
// fitness.GetMemberWorkloadWorstScore(sch),
// fitness.GetMemberWorkloadWorseScore(sch),
// fitness.GetMemberWorkloadBadScore(sch),
// fitness.GetPresidentSelfStudentScore(sch),
// fitness.GetSecretarySelfStudentScore(sch),
// fitness.GetExaminerNotPresidentScore(sch)
//};
/*foreach (FieldInfo info in typeof(Scores).GetFields().Where(x => x.IsStatic && x.IsLiteral))
* {
* results.Add((double)info.GetValue(info));
* //ws_info.Cells[row, 4].Value = info.Name;
* //ws_info.Cells[row, 5].Value = info.GetValue(info);
* //row++;
* results.Add()
*
* }*/
return(results.ToArray());
}
public Task <Schedule> RunAsync()
{
var selection = new EliteSelection();
var crossover = new UniformCrossover(0.5f);
//var mutation = new TworsMutation();
var mutation = new SchedulingMutation(ctx);
//var mutation = new UniformMutation();
var chromosome = new SchedulingChromosome(ctx);
Fitness = new SchedulingFitness(ctx);
var population = new Population(Parameters.MinPopulationSize, Parameters.MaxPopulationSize, chromosome);
termination = new SchedulingTermination();
geneticAlgorithm = new GeneticAlgorithm(population, Fitness, selection, crossover, mutation);
geneticAlgorithm.Termination = termination;
geneticAlgorithm.GenerationRan += GenerationRan;
geneticAlgorithm.MutationProbability = 0.05f;
return(Task.Run <Schedule>(
() =>
{
Console.WriteLine("GA running...");
geneticAlgorithm.Start();
Console.WriteLine("Best solution found has {0} fitness.", geneticAlgorithm.BestChromosome.Fitness);
var bestChromosome = geneticAlgorithm.BestChromosome as SchedulingChromosome;
var best = bestChromosome.Schedule;
return best;
}));
}
