private void btnRun_Click(object sender, EventArgs e)
{
double param1Value, param2Value, param3Value;
double[,] classprototypes;
int k;
SET_WAIT_CURSOR();
param1Value = (double)this.numParam1.Value;
param2Value = (double)this.numParam2.Value;
param3Value = (double)this.numParam3.Value;
if (DataSrc.HistoricalData.Constituents.Count > 0)
{
//If the population data is successfully retrieved
if (BuildPopulationMatrix())
{
//Run selected Algorithm
switch (AlgSelectList[algorithmSelectListBox.SelectedIndex])
{
case DefaultAlgorithm.K_MEANS:
if (param1Value > 0)
{
//Define the initial class prototypes as random points within feature space defined by the population
// this function limits the points to within a box defined by the min and max in each dimension plus _%
classprototypes = Helpers.XRandPointsInSpace((int)param1Value, PopulationToClassify);
k = Algorithms.k_means(PopulationToClassify, classprototypes, (int)param1Value, 5000, out PopulationClassLabels);
}
BindChartData(PopulationToClassify, PopulationClassLabels);
InsertClassColumn(PopulationClassLabels);
break;
case DefaultAlgorithm.FUZZY_C_MEANS:
if (param1Value > 0)
{
//Define the initial class prototypes as random points within feature space defined by the population
// this function limits the points to within a box defined by the min and max in each dimension plus _%
classprototypes = Helpers.XRandPointsInSpace((int)param1Value, PopulationToClassify);
k = Algorithms.fuzzy_c_means(PopulationToClassify, classprototypes, (int)param1Value, 2, 5000, out PopulationClassLabels);
}
BindChartData(PopulationToClassify, PopulationClassLabels);
InsertClassColumn(PopulationClassLabels);
break;
default:
break;
}
}
}
CLEAR_WAIT_CURSOR();
}
