static void Main(string[] argv)
{
modshogun.init_shogun_with_defaults();
int num = 1000;
double dist = 1.0;
double width = 2.1;
double C = 1.0;
DoubleMatrix offs =ones(2, num).mmul(dist);
DoubleMatrix x = randn(2, num).sub(offs);
DoubleMatrix y = randn(2, num).add(offs);
DoubleMatrix traindata_real = concatHorizontally(x, y);
DoubleMatrix m = randn(2, num).sub(offs);
DoubleMatrix n = randn(2, num).add(offs);
DoubleMatrix testdata_real = concatHorizontally(m, n);
DoubleMatrix o = ones(1,num);
DoubleMatrix trainlab = concatHorizontally(o.neg(), o);
DoubleMatrix testlab = concatHorizontally(o.neg(), o);
RealFeatures feats_train = new RealFeatures(traindata_real);
RealFeatures feats_test = new RealFeatures(testdata_real);
GaussianKernel kernel = new GaussianKernel(feats_train, feats_train, width);
Labels labels = new Labels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.train();
DoubleMatrix @out = svm.apply(feats_test).get_labels();
Console.WriteLine("Mean Error = " + signum(@out).ne(testlab).mean());
modshogun.exit_shogun();
}
static void Main(string[] argv)
{
modshogun.init_shogun_with_defaults();
double width = 2.1;
double epsilon = 1e-5;
double C = 1.0;
DoubleMatrix traindata_real = Load.load_numbers("../data/fm_train_real.dat");
DoubleMatrix testdata_real = Load.load_numbers("../data/fm_test_real.dat");
DoubleMatrix trainlab = Load.load_labels("../data/label_train_twoclass.dat");
RealFeatures feats_train = new RealFeatures();
feats_train.set_feature_matrix(traindata_real);
RealFeatures feats_test = new RealFeatures();
feats_test.set_feature_matrix(testdata_real);
GaussianKernel kernel = new GaussianKernel(feats_train, feats_train, width);
Labels labels = new Labels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.set_epsilon(epsilon);
svm.train();
kernel.init(feats_train, feats_test);
DoubleMatrix out_labels = svm.apply().get_labels();
Console.WriteLine(out_labels.ToString());
modshogun.exit_shogun();
}
public double[] ShapeClassification()
{
//Get single Test Instance from CSV file
CSVLoader loader = new CSVLoader();
loader.setSource(new java.io.File("GetOrientation.csv"));
Instances testinstances = loader.getDataSet();
testinstances.setClassIndex(testinstances.numAttributes() - 1);
Instance sekarang = testinstances.lastInstance();
//Get and build saved model
LibSVM modelShape = new LibSVM();
LibSVM modelOri = new LibSVM();
modelOri = (LibSVM)SerializationHelper.read("OrientationModel.model");
modelShape = (LibSVM)SerializationHelper.read("ShapeModel.model");
//Classify actual test instance
double valueShape = modelShape.classifyInstance(sekarang);
double valueOri = modelOri.classifyInstance(sekarang);
Console.WriteLine(valueOri);
Console.WriteLine(valueShape);
double[] value = new double[] { valueShape, valueOri };
return(value);
}
public static void Main()
{
modshogun.init_shogun_with_defaults();
double width = 2.1;
double epsilon = 1e-5;
double C = 1.0;
double[,] traindata_real = Load.load_numbers("../data/fm_train_real.dat");
double[,] testdata_real = Load.load_numbers("../data/fm_test_real.dat");
double[] trainlab = Load.load_labels("../data/label_train_twoclass.dat");
RealFeatures feats_train = new RealFeatures();
feats_train.set_feature_matrix(traindata_real);
RealFeatures feats_test = new RealFeatures();
feats_test.set_feature_matrix(testdata_real);
GaussianKernel kernel = new GaussianKernel(feats_train, feats_train, width);
BinaryLabels labels = new BinaryLabels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.set_epsilon(epsilon);
svm.train();
kernel.init(feats_train, feats_test);
double[] out_labels = LabelsFactory.to_binary(svm.apply()).get_labels();
foreach(double item in out_labels) {
Console.Write(item);
}
modshogun.exit_shogun();
}
public static void Main() {
modshogun.init_shogun_with_defaults();
int num = 1000;
double dist = 1.0;
double width = 2.1;
double C = 1.0;
Random RandomNumber = new Random();
double[,] traindata_real = new double[2, num * 2];
for (int i = 0; i < num; i ++) {
traindata_real[0, i] = RandomNumber.NextDouble() - dist;
traindata_real[0, i + num] = RandomNumber.NextDouble() + dist;
traindata_real[1, i] = RandomNumber.NextDouble() - dist;
traindata_real[1, i + num] = RandomNumber.NextDouble() + dist;
}
double[,] testdata_real = new double[2, num * 2];
for (int i = 0; i < num; i ++) {
testdata_real[0, i] = RandomNumber.NextDouble() - dist;
testdata_real[0, i + num] = RandomNumber.NextDouble() + dist;
testdata_real[1, i] = RandomNumber.NextDouble() - dist;
testdata_real[1, i + num] = RandomNumber.NextDouble() + dist;
}
double[] trainlab = new double[num * 2];
for (int i = 0; i < num; i ++) {
trainlab[i] = -1;
trainlab[i + num] = 1;
}
double[] testlab = new double[num * 2];
for (int i = 0; i < num; i ++) {
testlab[i] = -1;
testlab[i + num] = 1;
}
RealFeatures feats_train = new RealFeatures(traindata_real);
RealFeatures feats_test = new RealFeatures(testdata_real);
GaussianKernel kernel = new GaussianKernel(feats_train, feats_train, width);
BinaryLabels labels = new BinaryLabels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.train();
double[] result = LabelsFactory.to_binary(svm.apply(feats_test)).get_labels();
int err_num = 0;
for (int i = 0; i < num; i++) {
if (result[i] > 0) {
err_num += 1;
}
if (result[i+num] < 0) {
err_num += 1;
}
}
double testerr=err_num/(2*num);
Console.WriteLine(testerr);
}
static void Main(string[] argv)
{
modshogun.init_shogun_with_defaults();
double width = 2.1;
double epsilon = 1e-5;
double C = 1.0;
DoubleMatrix traindata_real = Load.load_numbers("../data/fm_train_real.dat");
DoubleMatrix testdata_real = Load.load_numbers("../data/fm_test_real.dat");
DoubleMatrix trainlab = Load.load_labels("../data/label_train_twoclass.dat");
CombinedKernel kernel = new CombinedKernel();
CombinedFeatures feats_train = new CombinedFeatures();
RealFeatures tfeats = new RealFeatures(traindata_real);
PolyKernel tkernel = new PolyKernel(10,3);
tkernel.init(tfeats, tfeats);
DoubleMatrix K = tkernel.get_kernel_matrix();
kernel.append_kernel(new CustomKernel(K));
RealFeatures subkfeats_train = new RealFeatures(traindata_real);
feats_train.append_feature_obj(subkfeats_train);
PolyKernel subkernel = new PolyKernel(10,2);
kernel.append_kernel(subkernel);
kernel.init(feats_train, feats_train);
Labels labels = new Labels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.train();
CombinedKernel kernel_pred = new CombinedKernel();
CombinedFeatures feats_pred = new CombinedFeatures();
RealFeatures pfeats = new RealFeatures(testdata_real);
PolyKernel tkernel_pred = new PolyKernel(10,3);
tkernel_pred.init(tfeats, pfeats);
DoubleMatrix KK = tkernel.get_kernel_matrix();
kernel_pred.append_kernel(new CustomKernel(KK));
RealFeatures subkfeats_test = new RealFeatures(testdata_real);
feats_pred.append_feature_obj(subkfeats_train);
PolyKernel subkernel_pred = new PolyKernel(10,2);
kernel_pred.append_kernel(subkernel_pred);
kernel_pred.init(feats_train, feats_pred);
svm.set_kernel(kernel_pred);
svm.apply();
DoubleMatrix km_train =kernel.get_kernel_matrix();
Console.WriteLine(km_train.ToString());
modshogun.exit_shogun();
}
public static void Main()
{
modshogun.init_shogun_with_defaults();
double width = 2.1;
double epsilon = 1e-5;
double C = 1.0;
double[,] traindata_real = Load.load_numbers("../data/fm_train_real.dat");
double[,] testdata_real = Load.load_numbers("../data/fm_test_real.dat");
double[] trainlab = Load.load_labels("../data/label_train_twoclass.dat");
RealFeatures feats_train = new RealFeatures();
feats_train.set_feature_matrix(traindata_real);
RealFeatures feats_test = new RealFeatures();
feats_test.set_feature_matrix(testdata_real);
GaussianKernel kernel = new GaussianKernel(feats_train, feats_train, width);
BinaryLabels labels = new BinaryLabels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.set_epsilon(epsilon);
svm.train();
kernel.init(feats_train, feats_test);
double[] out_labels = BinaryLabels.obtain_from_generic(svm.apply()).get_labels();
foreach (double item in out_labels)
{
Console.Write(item);
}
modshogun.exit_shogun();
}
public double MarginClassification()
{
//Get single Test Instance from CSV file
CSVLoader loader = new CSVLoader();
loader.setSource(new java.io.File("GetMargin.csv"));
Instances testinstances = loader.getDataSet();
testinstances.setClassIndex(testinstances.numAttributes() - 1);
Instance sekarang = testinstances.lastInstance();
//Get and build saved model
LibSVM svm = new LibSVM();
svm = (LibSVM)SerializationHelper.read("MarginModel.model");
//Classify actual test instance
double clsValue = svm.classifyInstance(sekarang);
Console.WriteLine(clsValue);
return(clsValue);
}
public Outcome MachineLearning(Instances instances)
{
Outcome outcome = new Outcome();
//load model
string projectPath = AppDomain.CurrentDomain.BaseDirectory;
LibSVM cls = (LibSVM)SerializationHelper.read(projectPath + "/MachineLearning/svmModel.model");
//predict outcome
instances.setClassIndex(19);
double[] values = cls.distributionForInstance(instances.instance(0));
if (values[0] < values[1])
{
outcome.Success = false;
}
else
{
outcome.Success = true;
}
return(outcome);
}
public static void testSVMPredictOnlyPackage()
{
string data_folder = @"D:\My_Dropbox\UProjects\OriBrainLearner\TestData\";
string data_file = data_folder + "svmguide1.t";
string range_file = data_folder + "svmguide1.range";
string svm_file = data_folder + "svmguide1.scale.model";
GuiPreferences.Instance.setLog("THIS PACKAGE ONLY PREDICTS!");
//Read in SVM model and feature range data
LibSVM svm = new LibSVM(); svm.LoadModel(svm_file);
double target_min, target_max; double[] features_min, features_max;
LibSVM.ReadRange(range_file, out target_min, out target_max, out features_min, out features_max);
//Read in Ground Truth data
List <int> labels_g = new List <int>();
List <LibSVMNode[]> samples = new List <LibSVMNode[]>();
using (StreamReader sr = new StreamReader(data_file))
{
string line; LibSVMNode[] sample; double label;
while ((line = sr.ReadLine()) != null)
{
LibSVM.ToLibSVMFormat(line, out sample, out label);
samples.Add(sample);
labels_g.Add((int)label);
}
}
//Convert to Array and List format
List <double[]> samples_array = new List <double[]>();
List <List <double> > samples_list = new List <List <double> >();
for (int i = 0; i < samples.Count; i++)
{
LibSVMNode[] sample = samples[i];
List <double> sample_list = new List <double>();
for (int k = 0, l = 0; k < sample.Length; k++)
{
int index = sample[k].index; if (index == -1)
{
break;
}
double value = sample[k].value;
for (int m = l; m < index - 1; m++)
{
sample_list.Add(double.NaN);
}
sample_list.Add(value); l = index;
}
samples_list.Add(sample_list);
samples_array.Add(sample_list.ToArray());
}
int count; double[] probs, dec_values;
//Scale the original data
LibSVMNode[][] samples_scaled = new LibSVMNode[samples.Count][];
GuiPreferences.Instance.setLog("Test scale data from SVMNode[] to SVMNode[]");
for (int i = 0; i < samples.Count; i++)
{
samples_scaled[i] = LibSVM.ScaleData(samples[i], target_min, target_max, features_min, features_max);
}
GuiPreferences.Instance.setLog("Test Predict functions. ");
count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.Predict(samples_scaled[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test scale data from double[] to SVMNode[]");
for (int i = 0; i < samples_array.Count; i++)
{
samples_scaled[i] = LibSVM.ScaleData(samples_array[i], target_min, target_max, features_min, features_max);
}
GuiPreferences.Instance.setLog("Test Predict functions. ");
count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.Predict(samples_scaled[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test scale data from list<double> to SVMNode[]");
for (int i = 0; i < samples_list.Count; i++)
{
samples_scaled[i] = LibSVM.ScaleData(samples_list[i], target_min, target_max, features_min, features_max);
}
GuiPreferences.Instance.setLog("Test Predict functions. ");
count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.Predict(samples_scaled[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test scale data from double[][] to SVMNode[][]");
samples_scaled = LibSVM.ScaleData(samples_array.ToArray(), target_min, target_max, features_min, features_max);
GuiPreferences.Instance.setLog("Test Predict functions. "); count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.Predict(samples_scaled[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test scale data from List<List<double>> to SVMNode[][]");
samples_scaled = LibSVM.ScaleData(samples_list, target_min, target_max, features_min, features_max);
GuiPreferences.Instance.setLog("Test Predict functions. "); count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.Predict(samples_scaled[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
//Convert scaled data to double[] and list<double> format
List <double[]> samples_scaled_array = new List <double[]>();
List <List <double> > samples_scaled_list = new List <List <double> >();
for (int i = 0; i < samples_scaled.Length; i++)
{
LibSVMNode[] sample = samples_scaled[i];
List <double> sample_list = new List <double>();
for (int k = 0, l = 0; k < sample.Length; k++)
{
int index = sample[k].index; if (index == -1)
{
break;
}
double value = sample[k].value;
for (int m = l; m < index - 1; m++)
{
sample_list.Add(double.NaN);
}
sample_list.Add(value); l = index;
}
samples_scaled_list.Add(sample_list);
samples_scaled_array.Add(sample_list.ToArray());
}
GuiPreferences.Instance.setLog("Test Predict(SVMNode[]) functions. "); count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.Predict(samples_scaled[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test Predict(List<double>) functions. "); count = 0;
for (int i = 0; i < samples_scaled_list.Count; i++)
{
double label = svm.Predict(samples_scaled_list[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test Predict(double[]) functions. "); count = 0;
for (int i = 0; i < samples_scaled_array.Count; i++)
{
double label = svm.Predict(samples_scaled_array[i]);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test PredictProb(SVMNode[]) functions. "); count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.PredictProb(samples_scaled[i], out probs);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test PredictProb(List<double>) functions. "); count = 0;
for (int i = 0; i < samples_scaled_list.Count; i++)
{
double label = svm.PredictProb(samples_scaled_list[i], out probs);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test PredictProb(double[]) functions. "); count = 0;
for (int i = 0; i < samples_scaled_array.Count; i++)
{
double label = svm.PredictProb(samples_scaled_array[i], out probs);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test PredictValues(SVMNode[]) functions. "); count = 0;
for (int i = 0; i < samples_scaled.GetLength(0); i++)
{
double label = svm.PredictValues(samples_scaled[i], out dec_values);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test PredictValues(List<double>) functions. "); count = 0;
for (int i = 0; i < samples_scaled_list.Count; i++)
{
double label = svm.PredictValues(samples_scaled_list[i], out dec_values);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
GuiPreferences.Instance.setLog("Test PredictValues(double[]) functions. "); count = 0;
for (int i = 0; i < samples_scaled_array.Count; i++)
{
double label = svm.PredictValues(samples_scaled_array[i], out dec_values);
if ((int)label == labels_g[i])
{
count++;
}
}
GuiPreferences.Instance.setLog("Accuracy = " + ((double)count / labels_g.Count).ToString());
}
public static void Main()
{
modshogun.init_shogun_with_defaults();
int num = 1000;
double dist = 1.0;
double width = 2.1;
double C = 1.0;
Random RandomNumber = new Random();
double[,] traindata_real = new double[2, num * 2];
for (int i = 0; i < num; i++)
{
traindata_real[0, i] = RandomNumber.NextDouble() - dist;
traindata_real[0, i + num] = RandomNumber.NextDouble() + dist;
traindata_real[1, i] = RandomNumber.NextDouble() - dist;
traindata_real[1, i + num] = RandomNumber.NextDouble() + dist;
}
double[,] testdata_real = new double[2, num * 2];
for (int i = 0; i < num; i++)
{
testdata_real[0, i] = RandomNumber.NextDouble() - dist;
testdata_real[0, i + num] = RandomNumber.NextDouble() + dist;
testdata_real[1, i] = RandomNumber.NextDouble() - dist;
testdata_real[1, i + num] = RandomNumber.NextDouble() + dist;
}
double[] trainlab = new double[num * 2];
for (int i = 0; i < num; i++)
{
trainlab[i] = -1;
trainlab[i + num] = 1;
}
double[] testlab = new double[num * 2];
for (int i = 0; i < num; i++)
{
testlab[i] = -1;
testlab[i + num] = 1;
}
RealFeatures feats_train = new RealFeatures(traindata_real);
RealFeatures feats_test = new RealFeatures(testdata_real);
GaussianKernel kernel = new GaussianKernel(feats_train, feats_train, width);
BinaryLabels labels = new BinaryLabels(trainlab);
LibSVM svm = new LibSVM(C, kernel, labels);
svm.train();
double[] result = BinaryLabels.obtain_from_generic(svm.apply(feats_test)).get_labels();
int err_num = 0;
for (int i = 0; i < num; i++)
{
if (result[i] > 0)
{
err_num += 1;
}
if (result[i + num] < 0)
{
err_num += 1;
}
}
double testerr = err_num / (2 * num);
Console.WriteLine(testerr);
modshogun.exit_shogun();
}
internal static HandleRef getCPtr(LibSVM obj)
{
return((obj == null) ? new HandleRef(null, IntPtr.Zero) : obj.swigCPtr);
}
internal static HandleRef getCPtr(LibSVM obj) {
return (obj == null) ? new HandleRef(null, IntPtr.Zero) : obj.swigCPtr;
}
