private void timer1_Tick(object sender, EventArgs e)
{
_lvq.LearnOneEpoch(_learningRate);
double d = (double)_lvq.Test() / _lvq.trainingData.Count;
nud_accuracy.Value = System.Convert.ToInt32((100 * d));
_graphics.Clear(Color.White);
pictureBox1.Image = _bitmap;
Drawing.DrawCodeVectors(_graphics, _lvq._w);
Drawing.DrawData(_graphics, _data, _width, _height);
}
static void Main(string[] args)
{
Random random = new Random();
string path = "iris.txt";
//string path = "diabetes.txt";
//string path = "vote.txt";
//string path = "iris2D.txt";
int N = 10; //k - fold
int M = 1; //liczba wektorów kodujących
List <List <Vector> > d = new List <List <Vector> >();
List <Vector> _data = DataFileReader.GetDataFromFile(path);
List <Vector> tempData = _data.ToList();
int b = _data.Count() / N;
for (int i = 0; i < N; i++)
{
d.Add(new List <Vector>());
for (int j = 0; j < b; j++)
{
int x = random.Next(tempData.Count);
d.Last().Add(tempData[x]);
tempData.RemoveAt(x);
}
}
d.Last().AddRange(tempData);
int num = 0;
for (int i = 0; i < N; i++)
{
LVQAlgorithm _lvq = new LVQAlgorithm();
_lvq.trainingData = makeTrainingSet(d, i);
_lvq.testingData = d[i];
_lvq.Initialize(M);
_lvq.Learn(100);
num += _lvq.Test();
}
Console.WriteLine("Crossvalidation {0}-folds", N);
Console.WriteLine("Number of Codebooks Vectors: {0}", M);
Console.WriteLine("Number of Instances : {0}", _data.Count);
Console.WriteLine("Correctly Classified Instances : {0}", num);
Console.WriteLine("Incorrectly Classified Instances : {0}", _data.Count - num);
Console.WriteLine("Accuracy: {0:F5}%", 100 * (double)num / _data.Count);
Console.ReadKey();
}
