Documentation of 'jsat.classifiers.linear.BBR' Java class
BBR
Skip navigation links
jsat.classifiers.linear

Class BBR

  • All Implemented Interfaces:
    java.io.Serializable, java.lang.Cloneable, Classifier, Parameterized, SimpleWeightVectorModel, SingleWeightVectorModel

    public class BBR
extends java.lang.Object
implements Classifier, Parameterized, SingleWeightVectorModel
    This is an implementation of Bayesian Binary Regression for L1 and L2 regularized logistic regression. This model requires additional memory to perform efficient column wise passes on the data set, assuming the data is sparse.

    BBR uses a Trust Region Newton algorithm that allows convergence to occur in a small number of iterations, but each iteration may be costly.

    See: Genkin, A., Lewis, D. D.,&Madigan, D. (2007). Large-Scale Bayesian Logistic Regression for Text Categorization. Technometrics, 49(3), 291–304. doi:10.1198/004017007000000245
    See Also:
    Serialized Form

    • Nested Class Summary

      Nested Classes 
      Modifier and Type Class and Description
      static class  BBR.Prior
      Valid priors that control what type of regularization is applied

    • Constructor Summary

      Constructors 
      Constructor and Description
      BBR(double regularization, int maxIterations)
      Creates a new BBR for L1 Logistic Regression object that will use the given regularization value.
      BBR(double regularization, int maxIterations, BBR.Prior prior)
      Creates a new BBR Logistic Regression object that will use the given regularization value.
      BBR(int maxIterations)
      Creates a new BBR for L1 Logistic Regression object that will attempt to automatically determine the regularization value to use.
      BBR(int maxIterations, BBR.Prior prior)
      Creates a new BBR Logistic Regression object that will attempt to automatically determine the regularization value to use.

    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method and Description
      CategoricalResults classify(DataPoint data)
      Performs classification on the given data point.
      BBR clone() 
      double getBias()
      Returns the bias term used for the model, or 0 of the model does not support or was not trained with a bias term.
      double getBias(int index)
      Returns the bias term used with the weight vector for the given class index.
      int getMaxIterations()
      Returns the maximum number of iterations allowed
      BBR.Prior getPrior()
      Returns the regularizing prior in use
      Vec getRawWeight()
      Returns the only weight vector used for the model
      Vec getRawWeight(int index)
      Returns the raw weight vector associated with the given class index.
      double getRegularization()
      Returns the regularization penalty used if the auto value is not used
      double getTolerance()
      Returns the tolerance parameter that controls convergence
      Vec getWeightVec()
      Returns the weight vector used to compute results via a dot product.
      boolean isAutoSetRegularization()
      Returns whether or not the algorithm will attempt to select the regularization term automatically
      boolean isUseBias()
      Returns true if a bias term is in use, false otherwise.
      int numWeightsVecs()
      Returns the number of weight vectors that can be returned.
      void setAutoSetRegularization(boolean autoSetRegularization)
      Sets whether or not the regularization term will be set automatically by the algorithm, which is done as specified in the original paper.
      void setMaxIterations(int maxIterations)
      Sets the maximum number of iterations allowed before halting the algorithm early.
      void setPrior(BBR.Prior prior)
      Sets the regularizing prior used
      void setRegularization(double regularization)
      Sets the regularization penalty to use if the algorithm has not been set to choose one automatically.
      void setTolerance(double tolerance)
      Sets the convergence tolerance target.
      void setUseBias(boolean useBias)
      Sets whether or not an implicit bias term should be added to the model.
      boolean supportsWeightedData()
      Indicates whether the model knows how to train using weighted data points.
      void train(ClassificationDataSet dataSet)
      Trains the classifier and constructs a model for classification using the given data set.
      void train(ClassificationDataSet dataSet, boolean parallel)
      Trains the classifier and constructs a model for classification using the given data set.

      • Methods inherited from class java.lang.Object

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

    • Constructor Detail

      • BBR

        public BBR(double regularization,
           int maxIterations)
        Creates a new BBR for L1 Logistic Regression object that will use the given regularization value.
        Parameters:
        regularization - the regularization penalty to apply
        maxIterations - the maximum number of training iterations to perform

      • BBR

        public BBR(double regularization,
           int maxIterations,
           BBR.Prior prior)
        Creates a new BBR Logistic Regression object that will use the given regularization value.
        Parameters:
        regularization - the regularization penalty to apply
        maxIterations - the maximum number of training iterations to perform
        prior - the prior to apply for regularization

      • BBR

        public BBR(int maxIterations)
        Creates a new BBR for L1 Logistic Regression object that will attempt to automatically determine the regularization value to use.
        Parameters:
        maxIterations - the maximum number of training iterations to perform

      • BBR

        public BBR(int maxIterations,
           BBR.Prior prior)
        Creates a new BBR Logistic Regression object that will attempt to automatically determine the regularization value to use.
        Parameters:
        maxIterations - the maximum number of training iterations to perform
        prior - the prior to apply for regularization

    • Method Detail

      • setRegularization

        public void setRegularization(double regularization)
        Sets the regularization penalty to use if the algorithm has not been set to choose one automatically.
        Parameters:
        regularization - sets the positive regularization penalty to use

      • getRegularization

        public double getRegularization()
        Returns the regularization penalty used if the auto value is not used
        Returns:
        the regularization penalty used if the auto value is not used

      • setAutoSetRegularization

        public void setAutoSetRegularization(boolean autoSetRegularization)
        Sets whether or not the regularization term will be set automatically by the algorithm, which is done as specified in the original paper. This may choose a very large (and bad) value of the regularization term, and should not be used with smaller data sets. This value is chosen deterministically.

        This value takes precedence over anything set with setRegularization(double)
        Parameters:
        autoSetRegularization - true to choose the regularization term automatically, false to use whatever value was set before.

      • isAutoSetRegularization

        public boolean isAutoSetRegularization()
        Returns whether or not the algorithm will attempt to select the regularization term automatically
        Returns:
        true if the regularization term is chosen automatically, false otherwise.

      • setMaxIterations

        public void setMaxIterations(int maxIterations)
        Sets the maximum number of iterations allowed before halting the algorithm early.
        Parameters:
        maxIterations - the maximum number of training iterations

      • getMaxIterations

        public int getMaxIterations()
        Returns the maximum number of iterations allowed
        Returns:
        the maximum number of iterations allowed

      • setTolerance

        public void setTolerance(double tolerance)
        Sets the convergence tolerance target. Relative changes that are smaller than the given tolerance will determine convergence.

        The default value used is that suggested in the original paper of 0.0005
        Parameters:
        tolerance - the positive convergence tolerance goal

      • getTolerance

        public double getTolerance()
        Returns the tolerance parameter that controls convergence
        Returns:
        the tolerance parameter that controls convergence

      • setUseBias

        public void setUseBias(boolean useBias)
        Sets whether or not an implicit bias term should be added to the model.
        Parameters:
        useBias - true to add a bias term, false to exclude the bias term.

      • isUseBias

        public boolean isUseBias()
        Returns true if a bias term is in use, false otherwise.
        Returns:
        true if a bias term is in use, false otherwise.

      • setPrior

        public void setPrior(BBR.Prior prior)
        Sets the regularizing prior used
        Parameters:
        prior - the prior to use

      • getPrior

        public BBR.Prior getPrior()
        Returns the regularizing prior in use
        Returns:
        the regularizing prior in use

      • getWeightVec

        public Vec getWeightVec()
        Returns the weight vector used to compute results via a dot product.
        Do not modify this value, or you will alter the results returned.
        Returns:
        the learned weight vector for prediction

      • getBias

        public double getBias()
        Description copied from interface: SingleWeightVectorModel
        Returns the bias term used for the model, or 0 of the model does not support or was not trained with a bias term.
        Specified by:
        getBias in interface SingleWeightVectorModel
        Returns:
        the bias term for the model

      • getRawWeight

        public Vec getRawWeight(int index)
        Description copied from interface: SimpleWeightVectorModel
        Returns the raw weight vector associated with the given class index. If the given class is an implicit zero vector, a ConstantVector object may be returned.
        Do not alter the returned weight vector, as it will change the model's values.

        If a regression problem, only index = 0 should be used
        Specified by:
        getRawWeight in interface SimpleWeightVectorModel
        Parameters:
        index - the class index to get the weight vector for
        Returns:
        the weight vector used for the specified class

      • getBias

        public double getBias(int index)
        Description copied from interface: SimpleWeightVectorModel
        Returns the bias term used with the weight vector for the given class index. If the model does not support or was not trained with bias weights, 0 will be returned.

        If a regression problem, only index = 0 should be used
        Specified by:
        getBias in interface SimpleWeightVectorModel
        Parameters:
        index - the class index to get the weight vector for
        Returns:
        the bias term for the specified class

      • numWeightsVecs

        public int numWeightsVecs()
        Description copied from interface: SimpleWeightVectorModel
        Returns the number of weight vectors that can be returned. For binary classification problems the value may be 1 if only a single weight vector's sign is used to determine the class. For multi-class problems, the weight vector count includes the implicit zero vector (if one is being used).
        Specified by:
        numWeightsVecs in interface SimpleWeightVectorModel
        Returns:
        the number of weight vectors for which SimpleWeightVectorModel.getRawWeight(int) can be called.

      • classify

        public CategoricalResults classify(DataPoint data)
        Description copied from interface: Classifier
        Performs classification on the given data point.
        Specified by:
        classify in interface Classifier
        Parameters:
        data - the data point to classify
        Returns:
        the results of the classification.

      • train

        public void train(ClassificationDataSet dataSet,
                  boolean parallel)
        Description copied from interface: Classifier
        Trains the classifier and constructs a model for classification using the given data set. If the training method knows how, it will used the threadPool to conduct training in parallel. This method will block until the training has completed.
        Specified by:
        train in interface Classifier
        Parameters:
        dataSet - the data set to train on
        parallel - true if multiple threads should be used to train the model. false if it should be done in a single threaded manner.

      • train

        public void train(ClassificationDataSet dataSet)
        Description copied from interface: Classifier
        Trains the classifier and constructs a model for classification using the given data set.
        Specified by:
        train in interface Classifier
        Parameters:
        dataSet - the data set to train on

      • supportsWeightedData

        public boolean supportsWeightedData()
        Description copied from interface: Classifier
        Indicates whether the model knows how to train using weighted data points. If it does, the model will train assuming the weights. The values returned by this method may change depending on the parameters set for the model.
        Specified by:
        supportsWeightedData in interface Classifier
        Returns:
        true if the model supports weighted data, false otherwise

      • clone

        public BBR clone()
        Specified by:
        clone in interface Classifier
        Overrides:
        clone in class java.lang.Object
Skip navigation links

DataMelt 3.0 © DataMelt by jWork.ORG

You see the box below because you did not login.