private static void Solve(string[] args)
{
IIoHandler inputManager = new IoHandler();
var data = inputManager.GetParameters(args);
IParser parser = new Parser(data);
Instance instance = parser.ParseData();
if (data.ExecuteAlgorithm)
{
GaSettings settings = parser.ParseSettings();
IAlgorithm algorithm = new Algorithm(instance, data.Time, settings);
Solution solution = algorithm.Solve(true);
parser.FormatAndSaveResult(solution);
}
else
{
ConsoleHandler.CheckResources(instance);
}
}
static Genome Start(string[] args)
{
InputData inputData = IoHandler.GetParameters(args);
FileHandler.ParseLines(inputData.FileName, out float[][] inputParams, out float[] desiredOutput);
Ga algorithm = null;
switch (inputData.Type)
{
case AlgorithmType.Generation:
algorithm = new GenerationGa(inputParams, desiredOutput, inputData);
break;
case AlgorithmType.Elimination:
algorithm = new EliminationGa(inputParams, desiredOutput, inputData);
break;
}
Genome bestSolution = algorithm?.Start();
return(bestSolution);
}
static void Start(string[] args)
{
InputData inputData = IoHandler.GetParameters(args);
IParser parser = new Parser(inputData);
GaSettings settings = parser.ParseSettings();
Ga algorithm;
switch (settings.Type)
{
case AlgorithmType.Generation:
algorithm = new GenerationGa(parser);
break;
case AlgorithmType.Elimination:
algorithm = new EliminationGa(parser);
break;
default:
throw new ArgumentOutOfRangeException();
}
Genome bestSolution = algorithm.Start();
Console.WriteLine("\n____________Proces is finished____________\n");
if (bestSolution != null)
{
Console.Write("Best result is: " + bestSolution.Fitness + ", with parameters: ");
PrintParameters(bestSolution.Genes);
}
Console.WriteLine("\n____________Classification____________\n");
algorithm.Test(out double[][] givenClasses, out double[][] expectedOutput);
bool[] correct = new bool[expectedOutput.Length];
for (int i = 0; i < expectedOutput.Length; i++)
{
correct[i] = true;
Console.Write("Expected classes: ");
for (int j = 0; j < expectedOutput[i].Length; j++)
{
Console.Write(expectedOutput[i][j].ToString("G0") + " ");
}
Console.Write("\tGiven classes: ");
for (int j = 0; j < givenClasses[i].Length; j++)
{
Console.Write(givenClasses[i][j].ToString("G0") + " ");
if (!(Math.Abs(expectedOutput[i][j] - givenClasses[i][j]) < double.Epsilon))
{
correct[i] = false;
}
}
Console.WriteLine(correct[i] ? "\tCorrect classification." : "\tIncorrect classification.");
}
int correctlyClassified = 0;
int incorrectlyClassified = 0;
for (int i = 0; i < correct.Length; i++)
{
if (correct[i])
{
correctlyClassified++;
}
else
{
incorrectlyClassified++;
}
}
Console.WriteLine("Correctly classified samples: " + correctlyClassified + ", incorrectly classified: " + incorrectlyClassified);
}