/// <summary>
/// Phần train cho K-SVMeans
/// </summary>
private void TrainKSVMeans(bool isBatchMode)
{
string strTrainFile = null;
if (isBatchMode)
{
strTrainFile = _trainFilePath;
}
else
{
strTrainFile = tbxTrainFilePath.Text;
}
int iNumCluster = (int)nmNumCluster.Value;
int iPos = strTrainFile.LastIndexOf('_');
string strMutualPath = strTrainFile.Remove(iPos + 1);
string strClusterModelFile = strMutualPath + "_clusterModel.txt";
string[] strClusterResultFiles = new string[iNumCluster];
string[] strSVMModelFiles = new string[iNumCluster];
for (int i = 0; i < iNumCluster; i++)
{
strClusterResultFiles[i] = strMutualPath + "cluster" + (i + 1).ToString() + ".txt";
strSVMModelFiles[i] = strMutualPath + "model" + (i + 1).ToString() + ".txt";
}
// Thực hiện cluster
SampleDataBUS samDataBUS = new SampleDataBUS();
samDataBUS.Read(strTrainFile);
Clustering clustering = new Clustering(iNumCluster, samDataBUS.Samples, DistanceType.Manhattan);
clustering.Run(strClusterModelFile, false);
samDataBUS.WriteIntoCluster(strClusterResultFiles, clustering.SampleData.ClusterIndices);
// Thực hiện train SVM
int iProgressBaseline = 0;
for (int i = 0; i < iNumCluster; i++)
{
Problem prob = Problem.Read(strClusterResultFiles[i]);
Parameter param = new Parameter();
iProgressBaseline = i * 100 / iNumCluster;
if (cmbModelSelection.SelectedItem.ToString() == "Grid search")
{
string strLogFile = strMutualPath + "GridCluster" + (i + 1).ToString() + ".txt";
double dblC;
double dblGamma;
ParameterSelection paramSel = new ParameterSelection();
paramSel.NFOLD = Int32.Parse(tbxNumFold.Text);
paramSel.EndEpochEvent += new CrossEpochEventHandler(
delegate(object senderNetwork, CrossEpochEventArgs args)
{
tlsProgressBar.Value = iProgressBaseline + (int)(args.TrainingIteration * 100d / (args.Cycles * iNumCluster));
tlsStatus.Text = "Cluster: " + (i + 1) + " | Current parameter set: " + args.TrainingIteration;
Application.DoEvents();
});
paramSel.Grid(prob, param, strLogFile, out dblC, out dblGamma);
param.C = dblC;
param.Gamma = dblGamma;
param.Probability = ckbProbEstimate.Checked;
Model model = Training.Train(prob, param);
Model.Write(strSVMModelFiles[i], model);
}
else if (cmbModelSelection.SelectedItem.ToString() == "Use default values")
{
if (tbxC.Text == "" || tbxGamma.Text == "")
{
MessageBox.Show("Please fill in parameters!");
return;
}
param.C = double.Parse(tbxC.Text);
param.Gamma = double.Parse(tbxGamma.Text);
param.Probability = ckbProbEstimate.Checked;
Model model = Training.Train(prob, param);
Model.Write(strSVMModelFiles[i], model);
}
}
tlsProgressBar.Value = 0;
}
private double[] doPrediction(double[] inputValue4Predicts, int numDayPredict,int modelType)
{
double[] result = null;
string strModelFile = null;
string strTestFile = null;
if (modelType == 0)//DT-ANN
{
strModelFile = _trainModel + "DT-ANN\\" + cmbStockID.SelectedItem.ToString()+ "_" + numDayPredict.ToString() + "_model.txt";
// Load model lên
BackpropagationNetwork bpNetwork;
Stream stream = File.Open(strModelFile, FileMode.Open);
BinaryFormatter bformatter = new BinaryFormatter();
bpNetwork = (BackpropagationNetwork)bformatter.Deserialize(stream);
stream.Close();
// Tạo tập mẫu để test
TrainingSample testSample = new TrainingSample(inputValue4Predicts);
// Thực hiện test
double[] dblTemp = bpNetwork.Run(testSample.InputVector);
result = new double[1];
result[0] = ConverterBUS.Convert2Trend(dblTemp);
}
else
{
//strModelFile = _trainModel + "K-SVMeans\\"; //+cmbStockID.SelectedItem.ToString() + "_" + numDayPredict.ToString() + "_model.txt";
string strMutualPath = _trainModel + "K-SVMeans\\" +cmbStockID.SelectedItem.ToString() + "_" + numDayPredict.ToString();
string strClusterModelFile = strMutualPath + "__clusterModel.txt";
// Thực hiện cluster
SampleDataBUS samDataBUS = new SampleDataBUS();
samDataBUS.Samples = new double[1][];
samDataBUS.Samples[0] = inputValue4Predicts;
//samDataBUS.Read(strTestFile);
Clustering clustering = new Clustering(samDataBUS.Samples, DistanceType.Manhattan);
clustering.Run(strClusterModelFile, true);
int clusterIndex = clustering.SampleData.ClusterIndices[0];
string strSVMModelFiles = strMutualPath + "_model" + (++clusterIndex).ToString() + ".txt";
Model model = Model.Read(strSVMModelFiles);
Node[] nodes = new Node[10];
for (int i = 0; i < 10; i++)
{
nodes[i] = new Node(i, inputValue4Predicts[i]);
}
double[] predictProbability = Prediction.PredictProbability(model, nodes);
//khoi tao label 3 pt
int[] labels = new int[3];
Procedures.svm_get_labels(model, labels);
result = new double[3];
for (int j = 0; j < labels.Length; j++)
{
if (labels[j] == 1)
{
result[0] = predictProbability[j];
}
else if (labels[j] == 0)
{
result[1] = predictProbability[j];
}
else
{
result[2] = predictProbability[j];
}
}
}
return result;
}
/// <summary>
/// Phần test cho K-SVMeans
/// </summary>
private void TestKSVMeans(bool isBatchMode)
{
string strModelFile = null;
string strTestFile = null;
if (isBatchMode)
{
strModelFile = _modelFilePath;
strTestFile = _testFilePath;
}
else
{
strModelFile = tbxModelFilePath.Text;
strTestFile = tbxTestFilePath.Text;
}
int iPos = strTestFile.LastIndexOf('_');
string strMutualPath = strTestFile.Remove(iPos + 1);
int iNumCluster = (int)nmNumCluster.Value;
string strClusterModelFile = strMutualPath + "_clusterModel.txt";
string[] strClusterResultFiles = new string[iNumCluster];
string[] strSVMModelFiles = new string[iNumCluster];
string[] strPredictedFiles = new string[iNumCluster];
string strStatisticFile = strMutualPath + "statistic.txt";
for (int i = 0; i < iNumCluster; i++)
{
strClusterResultFiles[i] = strMutualPath + "testcluster" + (i + 1).ToString() + ".txt";
strSVMModelFiles[i] = strMutualPath + "model" + (i + 1).ToString() + ".txt";
strPredictedFiles[i] = strMutualPath + "predict" + (i + 1).ToString() + ".txt";
}
// Thực hiện cluster
SampleDataBUS samDataBUS = new SampleDataBUS();
samDataBUS.Read(strTestFile);
Clustering clustering = new Clustering(samDataBUS.Samples, DistanceType.Manhattan);
clustering.Run(strClusterModelFile, true);
samDataBUS.WriteIntoCluster(strClusterResultFiles, clustering.SampleData.ClusterIndices);
// Thực hiện test SVM
StreamWriter writer = new StreamWriter(strMutualPath + "performance.txt");
double dblTotalPrecision = 0;
for (int i = 0; i < iNumCluster; i++)
{
Problem prob = Problem.Read(strClusterResultFiles[i]);
Model model = Model.Read(strSVMModelFiles[i]);
double dblPrecision = Prediction.Predict(prob, strPredictedFiles[i], model, ckbProbEstimate.Checked);
writer.WriteLine("Cluster " + (i + 1).ToString() + ": " + dblPrecision);
if (clustering.Clusters[i].NumSample > 0)
{
dblTotalPrecision += dblPrecision * clustering.Clusters[i].NumSample;
}
}
writer.WriteLine("All: " + dblTotalPrecision / samDataBUS.DataLines.Length);
writer.Close();
_resultShow += "<font color = \"blue\">Accuracy = " + (dblTotalPrecision / samDataBUS.DataLines.Length).ToString() + "%</font><br>";
StatisticTrend2File(strPredictedFiles, strStatisticFile);
}
