static void Main(string[] args)
{
SvmModelBuilder modelBuilder = new SvmModelBuilder();
string modelFileName = "model.svm";
if (!modelBuilder.LoadFromFile(modelFileName))
{
// first time usage, train from feature file
string problemFile = @"D:\UW\2012 Autumn\CSE 481\Kinect Capstone\WindowsFormsApplication1\WindowsFormsApplication1\rgbdfea_depth_first2.mat";
Console.WriteLine("Training new model from {0}.", problemFile);
modelBuilder.TrainModel(problemFile);
Console.WriteLine("Trainning finished. Saving Model as {0}", modelFileName);
modelBuilder.SaveModel(modelFileName);
}
Console.WriteLine("Model Ready.");
ImageFeature imgFeature = new ImageFeature();
short[,] imagePixels = new Png16Reader().Read("../../sampleImages/good/good_1/good_101_depthcrop.png");
double[] vector = imgFeature.GenerateFeature(imagePixels);
double category = svm.svm_predict(modelBuilder.GetModel(), MatrixUtil.DoubleToSvmNode(vector));
Console.WriteLine(category);
Console.ReadKey();
}
static void Main(string[] args)
{
SvmModelBuilder modelBuilder = new SvmModelBuilder();
string modelFileName = "model.svm";
if (!modelBuilder.LoadFromFile(modelFileName))
{
// first time usage, train from feature file
string problemFile = @"D:\UW\2012 Autumn\CSE 481\Kinect Capstone\WindowsFormsApplication1\WindowsFormsApplication1\rgbdfea_depth_first2.mat";
Console.WriteLine("Training new model from {0}.", problemFile);
modelBuilder.TrainModel(problemFile);
Console.WriteLine("Trainning finished. Saving Model as {0}", modelFileName);
modelBuilder.SaveModel(modelFileName);
}
Console.WriteLine("Model Ready.");
ImageFeature imgFeature = new ImageFeature();
short[,] imagePixels = new Png16Reader().Read("../../sampleImages/good/good_1/good_101_depthcrop.png");
double[] vector = imgFeature.GenerateFeature(imagePixels);
double category = svm.svm_predict(modelBuilder.GetModel(), MatrixUtil.DoubleToSvmNode(vector));
Console.WriteLine(category);
Console.ReadKey();
}
