edu.stanford.nlp.classify
Class AdaptedGaussianPriorObjectiveFunction<L,F>

java.lang.Object
  edu.stanford.nlp.optimization.AbstractCachingDiffFunction
      edu.stanford.nlp.optimization.AbstractStochasticCachingDiffFunction
          edu.stanford.nlp.optimization.AbstractStochasticCachingDiffUpdateFunction
              edu.stanford.nlp.classify.LogConditionalObjectiveFunction<L,F>
                  edu.stanford.nlp.classify.AdaptedGaussianPriorObjectiveFunction<L,F>

Type Parameters:

L - The type of the labels in the Dataset (one can be passed in to the constructor)

F - The type of the features in the Dataset

All Implemented Interfaces:

DiffFunction, Function, HasInitial

public class AdaptedGaussianPriorObjectiveFunction<L,F>
extends LogConditionalObjectiveFunction<L,F>

Adapt the mean of the Gaussian Prior by shifting the mean to the previously trained weights

Author:

Pi-Chuan Chang, Sarah Spikes (sdspikes@cs.stanford.edu) (Templatization)

Nested Class Summary

Nested classes/interfaces inherited from class edu.stanford.nlp.optimization.AbstractStochasticCachingDiffFunction

AbstractStochasticCachingDiffFunction.SamplingMethod

Field Summary

Fields inherited from class edu.stanford.nlp.classify.LogConditionalObjectiveFunction

data, dataIterable, dataweights, derivativeAD, derivativeNumerator, featureIndex, labelIndex, labels, numClasses, numFeatures, prior, priorDerivative, probs, sums, useIterable, useSummedConditionalLikelihood, values, xAD

Fields inherited from class edu.stanford.nlp.optimization.AbstractStochasticCachingDiffFunction

allIndices, curElement, finiteDifferenceStepSize, gradPerturbed, hasNewVals, HdotV, lastBatch, lastBatchSize, lastElement, lastVBatch, lastXBatch, method, randGenerator, recalculatePrevBatch, returnPreviousValues, sampleMethod, scaleUp, thisBatch, xPerturbed

Fields inherited from class edu.stanford.nlp.optimization.AbstractCachingDiffFunction

derivative, value

Constructor Summary

AdaptedGaussianPriorObjectiveFunction(GeneralDataset<L,F> dataset, LogPrior prior, double[][] weights)

Method Summary

protected void	calculate(double[] x) Calculate the conditional likelihood.
protected void	rvfcalculate(double[] x) Calculate conditional likelihood for datasets with real-valued features.
double[]	to1D(double[][] x2)

Methods inherited from class edu.stanford.nlp.classify.LogConditionalObjectiveFunction

calculateStochastic, calculateStochasticAlgorithmicDifferentiation, calculateStochasticFiniteDifference, calculateStochasticGradientOnly, calculateStochasticUpdate, dataDimension, domainDimension, indexOf, setPrior, setUseSumCondObjFun, to2D, valueAt

Methods inherited from class edu.stanford.nlp.optimization.AbstractStochasticCachingDiffUpdateFunction

calculateStochasticUpdate, getSample

Methods inherited from class edu.stanford.nlp.optimization.AbstractStochasticCachingDiffFunction

clearCache, decrementBatch, derivativeAt, derivativeAt, getBatch, HdotVAt, HdotVAt, HdotVAt, incrementBatch, incrementRandom, initial, lastDerivative, lastValue, scaleUp, valueAt, valueAt

Methods inherited from class edu.stanford.nlp.optimization.AbstractCachingDiffFunction

copy, derivativeAt, gradientCheck, randomInitial, valueAt

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

AdaptedGaussianPriorObjectiveFunction

public AdaptedGaussianPriorObjectiveFunction(GeneralDataset<L,F> dataset,
                                             LogPrior prior,
                                             double[][] weights)

Method Detail

calculate

protected void calculate(double[] x)

Calculate the conditional likelihood.

Overrides:

calculate in class LogConditionalObjectiveFunction<L,F>

Parameters:

x - The point at which to calculate the function

rvfcalculate

protected void rvfcalculate(double[] x)

Description copied from class: LogConditionalObjectiveFunction

Calculate conditional likelihood for datasets with real-valued features. Currently this can calculate CL only (no support for SCL). TODO: sum-conditional obj. fun. with RVFs.

Overrides:

rvfcalculate in class LogConditionalObjectiveFunction<L,F>

to1D

public double[] to1D(double[][] x2)