edu.stanford.nlp.classify
Class LogisticObjectiveFunction

java.lang.Object
  extended by edu.stanford.nlp.optimization.AbstractCachingDiffFunction
      extended by edu.stanford.nlp.classify.LogisticObjectiveFunction
All Implemented Interfaces:
DiffFunction, Function, HasInitial

public class LogisticObjectiveFunction
extends AbstractCachingDiffFunction

Maximizes the conditional likelihood with a given prior. Because the problem is binary, optimizations are possible that cannot be done in LogConditionalObjectiveFunction.

Author:
Galen Andrew

Field Summary
protected  float[] dataweights
           
 
Fields inherited from class edu.stanford.nlp.optimization.AbstractCachingDiffFunction
derivative, generator, value
 
Constructor Summary
LogisticObjectiveFunction(int numFeatures, int[][] data, double[][] values, int[] labels)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, double[][] values, int[] labels, float[] dataweights)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, double[][] values, int[] labels, LogPrior prior)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, double[][] values, int[] labels, LogPrior prior, float[] dataweights)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, int[] labels)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, int[] labels, float[] dataweights)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, int[] labels, LogPrior prior)
           
LogisticObjectiveFunction(int numFeatures, int[][] data, int[] labels, LogPrior prior, float[] dataweights)
           
 
Method Summary
protected  void calculate(double[] x)
          Calculate the value at x and the derivative and save them in the respective fields.
protected  void calculateRVF(double[] x)
           
 int domainDimension()
          Returns the number of dimensions in the function's domain
 
Methods inherited from class edu.stanford.nlp.optimization.AbstractCachingDiffFunction
clearCache, copy, derivativeAt, getDerivative, gradientCheck, gradientCheck, initial, lastValue, randomInitial, valueAt
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

dataweights

protected float[] dataweights
Constructor Detail

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 int[] labels)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 int[] labels,
                                 LogPrior prior)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 int[] labels,
                                 float[] dataweights)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 int[] labels,
                                 LogPrior prior,
                                 float[] dataweights)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 double[][] values,
                                 int[] labels)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 double[][] values,
                                 int[] labels,
                                 LogPrior prior)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 double[][] values,
                                 int[] labels,
                                 float[] dataweights)

LogisticObjectiveFunction

public LogisticObjectiveFunction(int numFeatures,
                                 int[][] data,
                                 double[][] values,
                                 int[] labels,
                                 LogPrior prior,
                                 float[] dataweights)
Method Detail

domainDimension

public int domainDimension()
Description copied from interface: Function
Returns the number of dimensions in the function's domain

Returns:
the number of domain dimensions

calculate

protected void calculate(double[] x)
Description copied from class: AbstractCachingDiffFunction
Calculate the value at x and the derivative and save them in the respective fields.

Specified by:
calculate in class AbstractCachingDiffFunction
Parameters:
x - The point at which to calculate the function

calculateRVF

protected void calculateRVF(double[] x)


Stanford NLP Group