Package edu.cmu.tetrad.bayes

Class MlBayesIm

java.lang.Object
edu.cmu.tetrad.bayes.MlBayesIm
All Implemented Interfaces:
BayesIm, Simulator, VariableSource, Im, TetradSerializable, Serializable
public final class MlBayesIm extends Object implements BayesIm
Stores a table of probabilities for a Bayes net and, together with BayesPm and Dag, provides methods to manipulate this table. The division of labor is as follows. The Dag is responsible for manipulating the basic graphical structure of the Bayes net. Dag also stores and manipulates the names of the nodes in the graph; there are no method in either BayesPm or BayesIm to do this. BayesPm stores and manipulates the *categories* of each node in a DAG, considered as a variable in a Bayes net. The number of categories for a variable can be changed there as well as the names for those categories. This class, BayesIm, stores the actual probability tables which are implied by the structures in the other two classes. The implied parameters take the form of conditional probabilities--e.g., P(N=v0|P1=v1, P2=v2, ...), for all nodes and all combinations of their parent categories. The set of all such probabilities is organized in this class as a three-dimensional table of double values. The first dimension corresponds to the nodes in the Bayes net. For each such node, the second dimension corresponds to a flat list of combinations of parent categories for that node. The third dimension corresponds to the list of categories for that node itself. Two methods allow these values to be set and retrieved:
  • getWordRatio(int nodeIndex, int rowIndex, int colIndex); and,
  • setProbability(int nodeIndex, int rowIndex, int colIndex, int probability).
To determine the index of the node in question, use the method
  • getNodeIndex(Node node).
To determine the index of the row in question, use the method
  • getRowIndex(int[] parentVals).
To determine the order of the parent values for a given node so that you can build the parentVals[] array, use the method
  • getParents(int nodeIndex)
To determine the index of a category, use the method
  • getCategoryIndex(Node node)
in BayesPm. The rest of the methods in this class are easily understood as variants of the methods above.

Thanks to Pucktada Treeratpituk, Frank Wimberly, and Willie Wheeler for advice and earlier versions.

Version:
$Id: $Id
Author:
josephramsey
See Also:

  • Field Summary

    Fields
    Modifier and Type
    Field
    Description
    static final int
    MANUAL
    Inidicates that new rows in this BayesIm should be initialized as unknowns, forcing them to be specified manually.
    static final int
    RANDOM
    Indicates that new rows in this BayesIm should be initialized randomly.

  • Constructor Summary

    Constructors
    Constructor
    Description
    MlBayesIm(BayesIm bayesIm)
    Copy constructor.
    MlBayesIm(BayesPm bayesPm)
    Constructs a new BayesIm from the given BayesPm, initializing all values as Double.NaN ("?").
    MlBayesIm(BayesPm bayesPm, int initializationMethod)
    Constructs a new BayesIm from the given BayesPm, initializing values either as MANUAL or RANDOM.
    MlBayesIm(BayesPm bayesPm, BayesIm oldBayesIm, int initializationMethod)
    Constructs a new BayesIm from the given BayesPm, initializing values either as MANUAL or RANDOM, but using values from the old BayesIm provided where posssible.

  • Method Summary

    Modifier and Type
    Method
    Description
    void
    clearRow(int nodeIndex, int rowIndex)
    Assigns random probability values to the child values of this row that add to 1.
    void
    clearTable(int nodeIndex)
    Randomizes every row in the table for the given node index.
    boolean
    equals(Object o)
    Returns true iff this Bayes net is equal to the given Bayes net.
    BayesPm
    getBayesPm()
    Getter for the field bayesPm.
    int
    getCorrespondingNodeIndex(int nodeIndex, BayesIm otherBayesIm)
    Returns the index of the given node in the given BayesIm.
    Graph
    getDag()
    getDag.
    List<Node>
    getMeasuredNodes()
    getMeasuredNodes.
    Node
    getNode(int nodeIndex)
    Returns the Node at the specified index.
    Node
    getNode(String name)
    getNode.
    int
    getNodeIndex(Node node)
    Gets the index of a given node within an array of nodes.
    int
    getNumColumns(int nodeIndex)
    Returns the number of columns of a node at the given index.
    int
    getNumNodes()
    getNumNodes.
    int
    getNumParents(int nodeIndex)
    Returns the number of parents of the given node.
    int
    getNumRows(int nodeIndex)
    Returns the number of rows in the provided node at the given index.
    static List<String>
    getParameterNames()
    getParameterNames.
    int
    getParent(int nodeIndex, int parentIndex)
    Returns the parent of a node given its index and parent index.
    int
    getParentDim(int nodeIndex, int parentIndex)
    Returns the dimension of the parent at the given node index and parent index.
    int[]
    getParentDims(int nodeIndex)
    Returns the dimensions of the pararents of the given node.
    int[]
    getParents(int nodeIndex)
    Returns the parents of the given node.
    int
    getParentValue(int nodeIndex, int rowIndex, int colIndex)
    Returns the given parent value.
    int[]
    getParentValues(int nodeIndex, int rowIndex)
    Returns the parents values of the given node.
    double
    getProbability(int nodeIndex, int rowIndex, int colIndex)
    Returns the probability for the given cell in the given CPT.
    int
    getRowIndex(int nodeIndex, int[] values)
    getRowIndex.
    List<String>
    getVariableNames()
    getVariableNames.
    List<Node>
    getVariables()
    getVariables.
    boolean
    isIncomplete(int nodeIndex)
    Returns true iff the given node has a Double.NaN value in it.
    boolean
    isIncomplete(int nodeIndex, int rowIndex)
    Returns true iff the given row in the given node has a Double.NaN value in it.
    void
    normalizeAll()
    Normalizes all rows in the tables associated with each of node in turn.
    void
    normalizeNode(int nodeIndex)
    Normalizes all rows in the table associated with a given node.
    void
    normalizeRow(int nodeIndex, int rowIndex)
    Normalizes the given row.
    void
    randomizeIncompleteRows(int nodeIndex)
    Randomizes any row in the table for the given node index that has a Double.NaN value in it.
    void
    randomizeRow(int nodeIndex, int rowIndex)
    Assigns random probability values to the child values of this row that add to 1.
    void
    randomizeTable(int nodeIndex)
    Randomizes every row in the table for the given node index.
    static MlBayesIm
    serializableInstance()
    Generates a simple exemplar of this class to test serialization.
    void
    setProbability(int nodeIndex, double[][] probMatrix)
    Sets the probability for the given node.
    void
    setProbability(int nodeIndex, int rowIndex, int colIndex, double value)
    Sets the probability for the given node at a given row and column in the table for that node.
    DataSet
    simulateData(int sampleSize, boolean latentDataSaved)
    Simulates a data set with the given number of rows and option to save latent variables.
    DataSet
    simulateData(int sampleSize, boolean latentDataSaved, int[] tiers)
    Simulates a sample with the given sample size.
    DataSet
    simulateData(DataSet dataSet, boolean latentDataSaved)
    Simulates data based on the provided data set and saves the latent data if specified.
    DataSet
    simulateData(DataSet dataSet, boolean latentDataSaved, int[] tiers)
    simulateData.
    String
    toString()
    Prints out the probability table for each variable.

    Methods inherited from class java.lang.Object

    getClass, hashCode, notify, notifyAll, wait, wait, wait

  • Field Details

    • MANUAL

      public static final int MANUAL
      Inidicates that new rows in this BayesIm should be initialized as unknowns, forcing them to be specified manually. This is the default.
      See Also:

    • RANDOM

      public static final int RANDOM
      Indicates that new rows in this BayesIm should be initialized randomly.
      See Also:

  • Constructor Details

    • MlBayesIm

      public MlBayesIm(BayesPm bayesPm) throws IllegalArgumentException
      Constructs a new BayesIm from the given BayesPm, initializing all values as Double.NaN ("?").
      Parameters:
      bayesPm - the given Bayes PM. Carries with it the underlying graph model.
      Throws:
      IllegalArgumentException - if the array of nodes provided is not a permutation of the nodes contained in the bayes parametric model provided.

    • MlBayesIm

      public MlBayesIm(BayesPm bayesPm, int initializationMethod) throws IllegalArgumentException
      Constructs a new BayesIm from the given BayesPm, initializing values either as MANUAL or RANDOM. If initialized manually, all values will be set to Double.NaN ("?") in each row; if initialized randomly, all values will distribute randomly in each row.
      Parameters:
      bayesPm - the given Bayes PM. Carries with it the underlying graph model.
      initializationMethod - either MANUAL or RANDOM.
      Throws:
      IllegalArgumentException - if the array of nodes provided is not a permutation of the nodes contained in the bayes parametric model provided.

    • MlBayesIm

      public MlBayesIm(BayesPm bayesPm, BayesIm oldBayesIm, int initializationMethod) throws IllegalArgumentException
      Constructs a new BayesIm from the given BayesPm, initializing values either as MANUAL or RANDOM, but using values from the old BayesIm provided where posssible. If initialized manually, all values that cannot be retrieved from oldBayesIm will be set to Double.NaN ("?") in each such row; if initialized randomly, all values that cannot be retrieved from oldBayesIm will distributed randomly in each such row.
      Parameters:
      bayesPm - the given Bayes PM. Carries with it the underlying graph model.
      oldBayesIm - an already-constructed BayesIm whose values may be used where possible to initialize this BayesIm. May be null.
      initializationMethod - either MANUAL or RANDOM.
      Throws:
      IllegalArgumentException - if the array of nodes provided is not a permutation of the nodes contained in the bayes parametric model provided.

    • MlBayesIm

      public MlBayesIm(BayesIm bayesIm) throws IllegalArgumentException
      Copy constructor.
      Parameters:
      bayesIm - a BayesIm object
      Throws:
      IllegalArgumentException - if any.

  • Method Details

    • serializableInstance

      public static MlBayesIm serializableInstance()
      Generates a simple exemplar of this class to test serialization.
      Returns:
      a MlBayesIm object

    • getParameterNames

      public static List<String> getParameterNames()

      getParameterNames.

      Returns:
      a List object

    • getBayesPm

      public BayesPm getBayesPm()

      Getter for the field bayesPm.

      Specified by:
      getBayesPm in interface BayesIm
      Returns:
      this PM.

    • getDag

      public Graph getDag()

      getDag.

      Specified by:
      getDag in interface BayesIm
      Returns:
      the DAG.

    • getNumNodes

      public int getNumNodes()

      getNumNodes.

      Specified by:
      getNumNodes in interface BayesIm
      Returns:
      the number of nodes in the model.

    • getNode

      public Node getNode(int nodeIndex)
      Returns the Node at the specified index.
      Specified by:
      getNode in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      Returns:
      the Node at the specified index.

    • getNode

      public Node getNode(String name)

      getNode.

      Specified by:
      getNode in interface BayesIm
      Parameters:
      name - the name of the node.
      Returns:
      the node.

    • getNodeIndex

      public int getNodeIndex(Node node)
      Gets the index of a given node within an array of nodes.
      Specified by:
      getNodeIndex in interface BayesIm
      Parameters:
      node - the given node.
      Returns:
      the index of the node in the array of nodes, or -1 if not found.

    • getVariables

      public List<Node> getVariables()

      getVariables.

      Specified by:
      getVariables in interface BayesIm
      Specified by:
      getVariables in interface VariableSource
      Returns:
      a List object

    • getMeasuredNodes

      public List<Node> getMeasuredNodes()

      getMeasuredNodes.

      Specified by:
      getMeasuredNodes in interface BayesIm
      Returns:
      the list of measured variableNodes.

    • getVariableNames

      public List<String> getVariableNames()

      getVariableNames.

      Specified by:
      getVariableNames in interface BayesIm
      Specified by:
      getVariableNames in interface VariableSource
      Returns:
      a List object

    • getNumColumns

      public int getNumColumns(int nodeIndex)
      Returns the number of columns of a node at the given index.
      Specified by:
      getNumColumns in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      Returns:
      the number of columns of the node at the given index.
      See Also:

    • getNumRows

      public int getNumRows(int nodeIndex)
      Returns the number of rows in the provided node at the given index.
      Specified by:
      getNumRows in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      Returns:
      the number of rows in the node.
      See Also:

    • getNumParents

      public int getNumParents(int nodeIndex)
      Returns the number of parents of the given node.
      Specified by:
      getNumParents in interface BayesIm
      Parameters:
      nodeIndex - the index of the node to get the number of parents for.
      Returns:
      the number of parents of the given node.

    • getParent

      public int getParent(int nodeIndex, int parentIndex)
      Returns the parent of a node given its index and parent index.
      Specified by:
      getParent in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      parentIndex - the index of the parent.
      Returns:
      the parent of the node.

    • getParentDim

      public int getParentDim(int nodeIndex, int parentIndex)
      Returns the dimension of the parent at the given node index and parent index.
      Specified by:
      getParentDim in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      parentIndex - the index of the parent.
      Returns:
      the dimension of the parent at the given node index and parent index.

    • getParentDims

      public int[] getParentDims(int nodeIndex)
      Returns the dimensions of the pararents of the given node.
      Specified by:
      getParentDims in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      Returns:
      (a defensive copy of) the array representing the dimensionality of each parent of a node, that is, the number of values which that node can take on. The order of entries in this array is the same as the order of entries of nodes returned by getParents() for that node.
      See Also:

    • getParents

      public int[] getParents(int nodeIndex)
      Returns the parents of the given node.
      Specified by:
      getParents in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      Returns:
      (a defensive copy of) the array containing all of the parents of a given node in the order in which they are stored internally.
      See Also:

    • getParentValues

      public int[] getParentValues(int nodeIndex, int rowIndex)
      Returns the parents values of the given node.
      Specified by:
      getParentValues in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      rowIndex - the index of the row in question.
      Returns:
      an array containing the combination of parent values for a given node and given row in the probability table for that node. To get the combination of parent values from the row number, the row number is represented using a variable-base place value system, where the bases for each place value are the dimensions of the parents in the order in which they are given by getParentDims(). For instance, if the row number (base 10) is 103 and the parent dimension array is [3 5 7], we calculate the first value as 103 / 7 = 14 with a remainder of 5. We then divide 14 / 5 = 2 with a remainder of 4. We then divide 2 / 3 = 0 with a remainder of 2. The variable place value representation is [2 4 5], which is the combination of parent values. This is the inverse function of getRowIndex().
      See Also:

    • getParentValue

      public int getParentValue(int nodeIndex, int rowIndex, int colIndex)
      Returns the given parent value.
      Specified by:
      getParentValue in interface BayesIm
      Parameters:
      nodeIndex - the index of the node.
      rowIndex - the index of the row in question.
      colIndex - the index of the column in question.
      Returns:
      the value in the probability table for the given node, at the given row and column.

    • getProbability

      public double getProbability(int nodeIndex, int rowIndex, int colIndex)
      Returns the probability for the given cell in the given CPT.
      Specified by:
      getProbability in interface BayesIm
      Parameters:
      nodeIndex - the index of the node in question.
      rowIndex - the row in the table for this for node which represents the combination of parent values in question.
      colIndex - the column in the table for this node which represents the value of the node in question.
      Returns:
      the probability for the given node at the given row and column in the table for that node. To get the node index, use getNodeIndex(). To get the row index, use getRowIndex(). To get the column index, use getCategoryIndex() from the underlying BayesPm(). The value returned will represent a conditional probability of the form P(N=v0 | P1=v1, P2=v2, ... , Pn=vn), where N is the node referenced by nodeIndex, v0 is the value referenced by colIndex, and the combination of parent values indicated is the combination indicated by rowIndex.
      See Also:

    • getRowIndex

      public int getRowIndex(int nodeIndex, int[] values)

      getRowIndex.

      Specified by:
      getRowIndex in interface BayesIm
      Parameters:
      nodeIndex - a int
      values - an array of
      invalid reference 
      int
      objects
      Returns:
      the row in the table for the given node and combination of parent values.
      See Also:

    • normalizeAll

      public void normalizeAll()
      Normalizes all rows in the tables associated with each of node in turn.
      Specified by:
      normalizeAll in interface BayesIm

    • normalizeNode

      public void normalizeNode(int nodeIndex)
      Normalizes all rows in the table associated with a given node.

      Normalizes all rows in the table associated with a given node.

      Specified by:
      normalizeNode in interface BayesIm
      Parameters:
      nodeIndex - the index of the node in question.

    • normalizeRow

      public void normalizeRow(int nodeIndex, int rowIndex)
      Normalizes the given row.

      Normalizes the given row.

      Specified by:
      normalizeRow in interface BayesIm
      Parameters:
      nodeIndex - the index of the node in question.
      rowIndex - the index of the row in question.

    • setProbability

      public void setProbability(int nodeIndex, double[][] probMatrix)
      Sets the probability for the given node. The matrix row represent row index, the row in the table for this for node which represents the combination of parent values in question. of the CPT. The matrix column represent column index, the column in the table for this node which represents the value of the node in question.

      Sets the probability for the given node. The matrix row represent row index, the row in the table for this for node which represents the combination of parent values in question. of the CPT. The matrix column represent column index, the column in the table for this node which represents the value of the node in question.

      Specified by:
      setProbability in interface BayesIm
      Parameters:
      nodeIndex - the index of the node in question.
      probMatrix - a matrix containing probabilities of a node along with its parents

    • setProbability

      public void setProbability(int nodeIndex, int rowIndex, int colIndex, double value)
      Sets the probability for the given node at a given row and column in the table for that node. To get the node index, use getNodeIndex(). To get the row index, use getRowIndex(). To get the column index, use getCategoryIndex() from the underlying BayesPm(). The value returned will represent a conditional probability of the form P(N=v0 | P1=v1, P2=v2, ... , Pn=vn), where N is the node referenced by nodeIndex, v0 is the value referenced by colIndex, and the combination of parent values indicated is the combination indicated by rowIndex.

      Sets the probability for the given node at a given row and column in the table for that node. To get the node index, use getNodeIndex(). To get the row index, use getRowIndex(). To get the column index, use getCategoryIndex() from the underlying BayesPm(). The value returned will represent a conditional probability of the form P(N=v0 | P1=v1, P2=v2, ... , Pn=vn), where N is the node referenced by nodeIndex, v0 is the value referenced by colIndex, and the combination of parent values indicated is the combination indicated by rowIndex.

      Specified by:
      setProbability in interface BayesIm
      Parameters:
      nodeIndex - the index of the node in question.
      rowIndex - the row in the table for this for node which represents the combination of parent values in question.
      colIndex - the column in the table for this node which represents the value of the node in question.
      value - the desired probability to be set.
      See Also:

    • getCorrespondingNodeIndex

      public int getCorrespondingNodeIndex(int nodeIndex, BayesIm otherBayesIm)
      Returns the index of the given node in the given BayesIm.
      Specified by:
      getCorrespondingNodeIndex in interface BayesIm
      Parameters:
      nodeIndex - the index of the node in this BayesIm.
      otherBayesIm - the BayesIm in which the node is to be found.
      Returns:
      the index of the node with the given name in the specified BayesIm.

    • clearRow

      public void clearRow(int nodeIndex, int rowIndex)
      Assigns random probability values to the child values of this row that add to 1.

      Assigns random probability values to the child values of this row that add to 1.

      Specified by:
      clearRow in interface BayesIm
      Parameters:
      nodeIndex - the node for the table that this row belongs to.
      rowIndex - the index of the row.

    • randomizeRow

      public void randomizeRow(int nodeIndex, int rowIndex)
      Assigns random probability values to the child values of this row that add to 1.

      Assigns random probability values to the child values of this row that add to 1.

      Specified by:
      randomizeRow in interface BayesIm
      Parameters:
      nodeIndex - the node for the table that this row belongs to.
      rowIndex - the index of the row.

    • randomizeIncompleteRows

      public void randomizeIncompleteRows(int nodeIndex)
      Randomizes any row in the table for the given node index that has a Double.NaN value in it.

      Randomizes any row in the table for the given node index that has a Double.NaN value in it.

      Specified by:
      randomizeIncompleteRows in interface BayesIm
      Parameters:
      nodeIndex - the node for the table whose incomplete rows are to be randomized.

    • randomizeTable

      public void randomizeTable(int nodeIndex)
      Randomizes every row in the table for the given node index.

      Randomizes every row in the table for the given node index.

      Specified by:
      randomizeTable in interface BayesIm
      Parameters:
      nodeIndex - the node for the table to be randomized.

    • clearTable

      public void clearTable(int nodeIndex)
      Randomizes every row in the table for the given node index.

      Randomizes every row in the table for the given node index.

      Specified by:
      clearTable in interface BayesIm
      Parameters:
      nodeIndex - the node for the table to be randomized.

    • isIncomplete

      public boolean isIncomplete(int nodeIndex, int rowIndex)
      Returns true iff the given row in the given node has a Double.NaN value in it.
      Specified by:
      isIncomplete in interface BayesIm
      Parameters:
      nodeIndex - the node for the table whose incomplete rows are to be checked.
      rowIndex - the index of the row in question.
      Returns:
      true iff one of the values in the given row is Double.NaN.

    • isIncomplete

      public boolean isIncomplete(int nodeIndex)
      Returns true iff the given node has a Double.NaN value in it.
      Specified by:
      isIncomplete in interface BayesIm
      Parameters:
      nodeIndex - the node for the table whose incomplete rows are to be checked.
      Returns:
      true iff any value in the table for the given node is Double.NaN.

    • simulateData

      public DataSet simulateData(int sampleSize, boolean latentDataSaved, int[] tiers)
      Simulates a sample with the given sample size.
      Parameters:
      sampleSize - the sample size.
      latentDataSaved - a boolean
      tiers - an array of
      invalid reference 
      int
      objects
      Returns:
      the simulated sample as a DataSet.

    • simulateData

      public DataSet simulateData(int sampleSize, boolean latentDataSaved)
      Simulates a data set with the given number of rows and option to save latent variables.
      Specified by:
      simulateData in interface BayesIm
      Specified by:
      simulateData in interface Simulator
      Parameters:
      sampleSize - the number of rows to simulate.
      latentDataSaved - if true, latent variables are saved in the data set.
      Returns:
      a simulated DataSet object.

    • simulateData

      public DataSet simulateData(DataSet dataSet, boolean latentDataSaved, int[] tiers)

      simulateData.

      Parameters:
      dataSet - a DataSet object
      latentDataSaved - a boolean
      tiers - an array of
      invalid reference 
      int
      objects
      Returns:
      a DataSet object

    • simulateData

      public DataSet simulateData(DataSet dataSet, boolean latentDataSaved)
      Simulates data based on the provided data set and saves the latent data if specified.
      Specified by:
      simulateData in interface BayesIm
      Parameters:
      dataSet - The data set to simulate.
      latentDataSaved - Indicates whether the latent data should be saved.
      Returns:
      The simulated data set.

    • equals

      public boolean equals(Object o)
      Returns true iff this Bayes net is equal to the given Bayes net. The sense of equality may vary depending on the type of Bayes net.
      Specified by:
      equals in interface BayesIm
      Overrides:
      equals in class Object
      Parameters:
      o - the Bayes net to be compared to this Bayes net.
      Returns:
      true iff this bayes net is equal to the given Bayes net. The sense of equality may vary depending on the type of Bayes net.

    • toString

      public String toString()
      Prints out the probability table for each variable.
      Specified by:
      toString in interface BayesIm
      Overrides:
      toString in class Object
      Returns:
      a String object