static public void SVMCrowdScaleForRegression(int trainNumber)
{
string folderName = trainNumber + "Regression";
SVMLearn svmLearn = new SVMLearn();
svmLearn.mode = Mode.Regression;
svmLearn.kernelType = Kernel.Polynomial;
svmLearn.ExecuteLearner("svm_learn.exe", folderName + "CrowdScale/train.dat", folderName + "CrowdScale/model.txt", folderName + "CrowdScale/learnLog.txt", false);
new SVMClassify().ExecuteClassifier("svm_classify.exe", folderName + "CrowdScale/test.dat", folderName + "CrowdScale/model.txt", folderName + "CrowdScale/output.txt", folderName + "CrowdScale/clasifylog.txt", false);
}
static private void SVMExample()
{
SVMLearn svmLearn = new SVMLearn();
svmLearn.ExecuteLearner("svm_learn.exe", "example/train.dat", "example/model.txt", "example/learnLog.txt", false);
SVMClassify svmClassify = new SVMClassify();
svmClassify.ExecuteClassifier("svm_classify.exe", "example/test.dat", "example/model.txt", "example/output.txt", "example/clasifylog.txt", false);
}
static public void SVMCrowdScale(int numberOfTrain, Kernel kernel, int ParamD, double ParamG, double ParamS, double ParamC)
{
string folderName = numberOfTrain + "Classification";
SVMLearn svmLearn0 = new SVMLearn();
svmLearn0.kernelType = kernel;
svmLearn0.ParamD = ParamD;
svmLearn0.ParamG = ParamG;
svmLearn0.ParamS = ParamS;
svmLearn0.ParamC = ParamC;
svmLearn0.ExecuteLearner("svm_learn.exe", folderName + "CrowdScale/0train.dat", folderName + "CrowdScale/0model.txt", folderName + "CrowdScale/0learnLog.txt", false);
new SVMClassify().ExecuteClassifier("svm_classify.exe", folderName + "CrowdScale/0test.dat", folderName + "CrowdScale/0model.txt", folderName + "CrowdScale/0output.txt", folderName + "CrowdScale/0clasifylog.txt", false);
SVMLearn svmLearn1 = new SVMLearn();
svmLearn1.kernelType = kernel;
svmLearn1.ParamD = ParamD;
svmLearn1.ParamG = ParamG;
svmLearn1.ParamS = ParamS;
svmLearn1.ParamC = ParamC;
svmLearn1.ExecuteLearner("svm_learn.exe", folderName + "CrowdScale/1train.dat", folderName + "CrowdScale/1model.txt", folderName + "CrowdScale/1learnLog.txt", false);
new SVMClassify().ExecuteClassifier("svm_classify.exe", folderName + "CrowdScale/1test.dat", folderName + "CrowdScale/1model.txt", folderName + "CrowdScale/1output.txt", folderName + "CrowdScale/1clasifylog.txt", false);
SVMLearn svmLearn2 = new SVMLearn();
svmLearn2.kernelType = kernel;
svmLearn2.ParamD = ParamD;
svmLearn2.ParamG = ParamG;
svmLearn2.ParamS = ParamS;
svmLearn2.ParamC = ParamC;
svmLearn2.ExecuteLearner("svm_learn.exe", folderName + "CrowdScale/2train.dat", folderName + "CrowdScale/2model.txt", folderName + "CrowdScale/2learnLog.txt", false);
new SVMClassify().ExecuteClassifier("svm_classify.exe", folderName + "CrowdScale/2test.dat", folderName + "CrowdScale/2model.txt", folderName + "CrowdScale/2output.txt", folderName + "CrowdScale/2clasifylog.txt", false);
SVMLearn svmLearn3 = new SVMLearn();
svmLearn3.kernelType = kernel;
svmLearn3.ParamD = ParamD;
svmLearn3.ParamG = ParamG;
svmLearn3.ParamS = ParamS;
svmLearn3.ParamC = ParamC;
svmLearn3.ExecuteLearner("svm_learn.exe", folderName + "CrowdScale/3train.dat", folderName + "CrowdScale/3model.txt", folderName + "CrowdScale/3learnLog.txt", false);
new SVMClassify().ExecuteClassifier("svm_classify.exe", folderName + "CrowdScale/3test.dat", folderName + "CrowdScale/3model.txt", folderName + "CrowdScale/3output.txt", folderName + "CrowdScale/3clasifylog.txt", false);
SVMLearn svmLearn4 = new SVMLearn();
svmLearn4.kernelType = kernel;
svmLearn4.ParamD = ParamD;
svmLearn4.ParamG = ParamG;
svmLearn4.ParamS = ParamS;
svmLearn4.ParamC = ParamC;
svmLearn4.ExecuteLearner("svm_learn.exe", folderName + "CrowdScale/4train.dat", folderName + "CrowdScale/4model.txt", folderName + "CrowdScale/4learnLog.txt", false);
new SVMClassify().ExecuteClassifier("svm_classify.exe", folderName + "CrowdScale/4test.dat", folderName + "CrowdScale/4model.txt", folderName + "CrowdScale/4output.txt", folderName + "CrowdScale/4clasifylog.txt", false);
}
public override void buildClassifier(Instances instances)
{
m_mustValue = null;
var weights = MincostLiblinearClassifier.GetCount(instances);
if (weights[0] == 0)
{
m_mustValue = 2;
m_delta = 0;
return;
}
else if (weights[2] == 0)
{
m_mustValue = 0;
m_delta = 0;
return;
}
m_sampleInstances = new Instances(instances, 0);
// can classifier handle the data?
getCapabilities().testWithFail(instances);
Instances trainInstances = new Instances(instances, 0, instances.numInstances());
AddInstancesAccordWeight(trainInstances);
if (System.IO.File.Exists(m_trainingFile))
{
System.IO.File.Delete(m_trainingFile);
}
libsvmSaver.setInstances(trainInstances);
libsvmSaver.setFile(new java.io.File(m_trainingFile));
libsvmSaver.writeBatch();
//ConvertNorminalToString(m_trainingFile);
if (System.IO.File.Exists(m_modelFile))
{
System.IO.File.Delete(m_modelFile);
}
string[] options = Utils.splitOptions(m_trainArgs);
int idx = Utils.getOptionPos('c', options);
if (idx != -1)
{
double c = Convert.ToDouble(options[idx + 1]);
c = c * trainInstances.numInstances() / 100.0;
options[idx + 1] = c.ToString();
m_trainArgs = Utils.joinOptions(options);
}
learner.ExecuteLearner(s_learnerPath, m_trainingFile, m_modelFile, m_trainArgs);
if (!System.IO.File.Exists(m_modelFile))
{
throw new InvalidOperationException(learner.Output);
}
m_modelData = System.IO.File.ReadAllBytes(m_modelFile);
GetBestDelta(instances);
if (System.IO.File.Exists(m_trainingFile))
{
System.IO.File.Delete(m_trainingFile);
}
}