//Experiment
public void GenerateExperiments()
{
int[] trainQua = { 4, 40, 400, 4000 };
int[] testQua = { 4, 40, 400, 4000 };
List <string> rowsOrg = new List <string>();
List <string> rowsLib = new List <string>();
AddHeaders(rowsOrg);
CultureInfo lng = new CultureInfo("en-US");
for (int i = 0; i < trainQua.Length; i++)
{
for (int j = 0; j < testQua.Length; j++)
{
for (int k = 0; k < TREATMENT_REP; k++)
{
DataTable[] tables = GenerateExperimentDataTables(trainQua[i], testQua[j]);
//Original
DecisionTree.Model.DecisionTree original = new DecisionTree.Model.DecisionTree(tables[0]);
double perOrg = original.Test(tables[1]);
//...
//Library
Codification codebook = GenerateDecisionTreeLib(tables[0]);
double perLib = DecisionTreeAccuracyPercentageLib(tables[1], codebook);
//...
rowsOrg.Add("Original" + "," + trainQua[i] + "," + testQua[j] + "," + (k + 1) + "," + perOrg.ToString("0.#####", lng));
rowsLib.Add("Library" + "," + trainQua[i] + "," + testQua[j] + "," + (k + 1) + ", " + perLib.ToString("0.#####", lng));
LoadedData++;
ActualLine = "#Train Quantity:" + trainQua[i] + " #Test Quantity:" + testQua[j] + " #Repetition: " + (k + 1);
}
}
}
ExportResults(rowsOrg, rowsLib);
}
//...
//Machine Learning
//Original
public void GenerateDecisionTreeOrg()
{
decisionTreeOrg = new DecisionTree.Model.DecisionTree(GenerateTrainingDataTableOrg());
}
