binaryprediction.java

Boosting算法软件包

package jboost.booster;

import jboost.examples.Label;

/**
 * This is a prediction for a binary label, which consists of a single
 * real valued number whose sign is the prediction and whose magnitude
 * is the prediction confidence
 **/
class BinaryPrediction extends Prediction{

    protected double prediction;
    public BinaryPrediction(double p) {prediction=p;}
    public BinaryPrediction() {prediction=0.0;}
    
    public Object clone(){
	Object a = new BinaryPrediction(prediction);
	return a;
    }

    public Prediction add(Prediction p) {
	prediction += ((BinaryPrediction) p).prediction;
	return this;
    }

    public Prediction scale(double w) {
	prediction *= w;
	return this;
    }
    
    public Prediction add(double w, Prediction p) {
	prediction += w * ((BinaryPrediction) p).prediction;
	return this;
    }
    
    public double[] getClassScores() {
	double[] a = {-prediction,prediction};
	return a;
    }

    /**
     * computes margin as y * prediction, where y = -1 if label = 0,
     * y = +1 if label = 1.
     */
    public double[] getMargins(Label label) {
	int l = label.getSingleValue();
	if(l > 1 || l < 0) {
	    throw new IllegalArgumentException(
		       "Adaboost.getMargin should get a label which " +
		       "is either 0 or 1. Instead it got " + l);
	}
	return new double[] { (l == 1 ? prediction : -prediction) };
    }

    public boolean equals(Prediction other) {
      BinaryPrediction bp= (BinaryPrediction) other;
      return (prediction == bp.prediction); 
    }
    
    /**
     * computes margin as absolute value of prediction.
     */
    /*
    public double getMargin() {
	return (prediction < 0.0 ? -prediction : prediction);
    }
    */

    public Label getBestClass() {
	return new Label(prediction > 0 ? 1 : 0);
    }

    public String toString() {
	return "BinaryPrediction. p(1)= "+prediction;
    }

    public String shortText() {
	return Double.toString(prediction);
    }

    public String cPreamble() {
      return
        "typedef double Prediction_t;\n" +
        "#define reset_pred()  {p = 0.0;}\n" +
	"#define add_pred(X) {p += (X);}\n" +
        "#define finalize_pred()" +
            "  ((r) ? (r[1] = p , r[0] = -p) : -p)\n";
    }

    public String javaPreamble() {
      return ""
	  + "  static private double p;\n"
	  + "  static private void reset_pred() { p = 0.0; }\n"
	  + "  static private void add_pred(double x) { p += x; }\n"
	  + "  static private double[] finalize_pred() {\n"
	  + "    return new double[] {-p, p};\n"
	  + "  }\n";
    }

    public double[] toCodeArray() {
	return new double[] {prediction};
    }

}