/// <summary>
    ///   Initializes a new instance of a Sequential Minimal Optimization (SMO) algorithm.
    /// </summary>
    /// 
    /// <param name="machine">A Support Vector Machine.</param>
    /// <param name="inputs">The input data points as row vectors.</param>
    /// <param name="outputs">The classification label for each data point in the range [-1;+1].</param>
    /// 
    public SequentialMinimalOptimization(SupportVectorMachine machine,
                                         double[][] inputs, int[] outputs)
    {

      // Initial argument checking
      if (machine == null)
        throw new ArgumentNullException("machine");

      if (inputs == null)
        throw new ArgumentNullException("inputs");

      if (outputs == null)
        throw new ArgumentNullException("outputs");

      if (inputs.Length != outputs.Length)
        throw new ArgumentException("The number of inputs and outputs does not match.", "outputs");

      for (int i = 0; i < outputs.Length; i++)
      {
        if (outputs[i] != 1 && outputs[i] != -1)
          throw new ArgumentOutOfRangeException("outputs", "One of the labels in the output vector is neither +1 or -1.");
      }

      if (machine.Inputs > 0)
      {
        // This machine has a fixed input vector size
        for (int i = 0; i < inputs.Length; i++)
          if (inputs[i].Length != machine.Inputs)
            throw new ArgumentException(
              "The size of the input vectors does not match the expected number of inputs of the machine");
      }


      // Machine
      this.machine = machine;

      // Kernel (if applicable)
      KernelSupportVectorMachine ksvm = machine as KernelSupportVectorMachine;
      this.kernel = (ksvm != null) ? ksvm.Kernel : new Linear();


      // Learning data
      this.inputs = inputs;
      this.outputs = outputs;

    }
示例#2
0
    public void CreateKernelplot(ZedGraphControl zgc, double[,] graph, SupportVectorMachine svm)
    {
      GraphPane myPane = zgc.GraphPane;
      myPane.CurveList.Clear();

      // Set the titles
      myPane.Title.IsVisible = false;
      myPane.XAxis.Title.Text = _sourceColumns[0];
      myPane.YAxis.Title.Text = _sourceColumns[1];


      // Classification problem
      PointPairList list1 = new PointPairList(); // Z = -1
      PointPairList list2 = new PointPairList(); // Z = +1
      for (int i = 0; i < graph.GetLength(0); i++)
      {
        if (graph[i, 2] == -1)
          list1.Add(graph[i, 0], graph[i, 1]);
        if (graph[i, 2] == 1)
          list2.Add(graph[i, 0], graph[i, 1]);
      }

      var list3 = new PointPairList();
      var list_up = new PointPairList();
      var list_down = new PointPairList();
      double x, y = 0.0, y_down = 0.0, y_up = 0.0;
      for (var i = -2; i < 12; i++)
      {
        var a = -svm.Weights[0] / svm.Weights[1];
        y = a * i - svm.Threshold / svm.Weights[1];
        var b = svm.SupportVectorsTwoSides[0];
        y_up = a * i + (b[1] - a * b[0]);
        list3.Add(i, y);
        list_up.Add(i, y_up);

        b = svm.SupportVectorsTwoSides[1];
        y_down = a * i + (b[1] - a * b[0]);
        list_down.Add(i, y_down);
      }


      // Add the curve
      LineItem myCurve = myPane.AddCurve("G1", list1, Color.Blue, SymbolType.Diamond);
      myCurve.Line.IsVisible = false;
      myCurve.Symbol.Border.IsVisible = false;
      myCurve.Symbol.Fill = new Fill(Color.Blue);

      myCurve = myPane.AddCurve("G2", list2, Color.Green, SymbolType.Diamond);
      myCurve.Line.IsVisible = false;
      myCurve.Symbol.Border.IsVisible = false;
      myCurve.Symbol.Fill = new Fill(Color.Green);

      myCurve = myPane.AddCurve("L", list3, Color.Black, SymbolType.Circle);
      myCurve.Line.IsVisible = true;

      myCurve = myPane.AddCurve("L", list_up, Color.Black, SymbolType.Circle);
      myCurve.Line.IsVisible = true;

      myCurve = myPane.AddCurve("L", list_down, Color.Black, SymbolType.Circle);
      myCurve.Line.IsVisible = true;

      // Fill the background of the chart rect and pane
      //myPane.Chart.Fill = new Fill(Color.White, Color.LightGoldenrodYellow, 45.0f);
      //myPane.Fill = new Fill(Color.White, Color.SlateGray, 45.0f);
      myPane.Fill = new Fill(Color.WhiteSmoke);

      zgc.AxisChange();
      zgc.Invalidate();
    }
示例#3
0
    private void btnSimpleSMO_Click(object sender, EventArgs e)
    {
      double[,] sourceMatrix;
      double[,] inputs;
      int[] labels;

      GetData(out sourceMatrix, out inputs, out labels);

      //_svm.SimpleSMO(inputs, labels);
      // Perform classification
      SequentialMinimalOptimization smo;

      // Creates the Support Vector Machine using the selected kernel
      //svm = new KernelSupportVectorMachine(kernel, 2);
      SupportVectorMachine svm = new SupportVectorMachine(2);

      // Creates a new instance of the SMO Learning Algorithm
      smo = new SequentialMinimalOptimization(svm, inputs.ToArray(), labels);

      // Set learning parameters
      smo.Complexity =1.0;
      smo.Tolerance = 0.001;


      bool converged = true;

      try
      {
        // Run
        double error = smo.Run();
      }
      catch 
      {
        converged = false;
      }

      var a = sourceMatrix.Submatrix(0, sourceMatrix.GetLength(0) - 1, 0, 2);
      CreateScatterplot(graphInput, a, svm);
    }