org.apache.mahout.classifier.discriminative
Class LinearModel

java.lang.Object
  extended by org.apache.mahout.classifier.discriminative.LinearModel

public class LinearModel
extends java.lang.Object

Classifies a data point using a hyperplane.


Constructor Summary
LinearModel(Vector hyperplane)
          Init a linear model with zero displacement and a threshold of 0.5.
LinearModel(Vector hyperplane, double displacement, double threshold)
          Init a linear model with a hyperplane, distance and displacement.
 
Method Summary
 void addDelta(Vector delta)
          Update the hyperplane by adding delta.
 boolean classify(Vector dataPoint)
          Classify a point to either belong to the class modeled by this linear model or not.
 void shiftBias(double factor)
          Shift the bias of the model.
 void timesDelta(int index, double delta)
          Multiply the weight at index by delta.
 java.lang.String toString()
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait
 

Constructor Detail

LinearModel

public LinearModel(Vector hyperplane,
                   double displacement,
                   double threshold)
Init a linear model with a hyperplane, distance and displacement.


LinearModel

public LinearModel(Vector hyperplane)
Init a linear model with zero displacement and a threshold of 0.5.

Method Detail

classify

public boolean classify(Vector dataPoint)
Classify a point to either belong to the class modeled by this linear model or not.

Parameters:
dataPoint - the data point to classify.
Returns:
returns true if data point should be classified as belonging to this model.

addDelta

public void addDelta(Vector delta)
Update the hyperplane by adding delta.

Parameters:
delta - the delta to add to the hyperplane vector.

toString

public java.lang.String toString()
Overrides:
toString in class java.lang.Object

shiftBias

public void shiftBias(double factor)
Shift the bias of the model.

Parameters:
factor - factor to multiply the bias by.

timesDelta

public void timesDelta(int index,
                       double delta)
Multiply the weight at index by delta.

Parameters:
index - the index of the element to update.
delta - the delta to multiply the element with.


Copyright © 2008-2010 The Apache Software Foundation. All Rights Reserved.