Documentation of 'com.datumbox.framework.core.common.dataobjects.Record' Java class
Record
com.datumbox.framework.core.common.dataobjects

Class Record

  • All Implemented Interfaces:
    java.io.Serializable


    public class Record
    extends java.lang.Object
    implements java.io.Serializable
    The Record is the object that stores all the entries of the Dataset and it contains information about the x, y, predicted class and probabilities of the predictions. The record is immutable. In general Machine Learning methods can work with a variety of different types (Numbers, Boolean, Categorical and Ordinal). The types of the provided data are not known at compile time but rather on runtime and they heavily depend on the case or method. As a result all the data of x and y in the Record object, are stored internally as Objects. Moreover note that the column names of X are also defined as Object. This is because depending on the case, the columns can be Strings, Integers or even Tuples in the case of categorical features (tuples are implemented as immutable List<Object>). For the response variable y it applies the same; depending on the method it can be a Number, Boolean, Categorical or Ordinal. As a result it is also stored as Object.
    See Also:
    Serialized Form
    • Constructor Summary

      Constructors 
      Constructor and Description
      Record(AssociativeArray x, java.lang.Object y)
      Constructor which takes as arguments the x and y data.
      Record(AssociativeArray x, java.lang.Object y, java.lang.Object yPredicted, AssociativeArray yPredictedProbabilities)
      Constructor which takes as arguments the x, y, predicted y and predicted probaibilites.
    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method and Description
      boolean equals(java.lang.Object obj)
      Checks if the provided object is equal to the current instance.
      AssociativeArray getX()
      It returns an AssociativeArray with all the xData of the Record.
      java.lang.Object getY()
      It returns the real response variable of the Record.
      java.lang.Object getYPredicted()
      It returns the predicted response variable of the Record.
      AssociativeArray getYPredictedProbabilities()
      If prediction probabilities are available for the particular algorithm, then this method will return these probabilities.
      int hashCode()
      The hash code of the record.
      java.lang.String toString()
      • Methods inherited from class java.lang.Object

        getClass, notify, notifyAll, wait, wait, wait
    • Constructor Detail

      • Record

        public Record(AssociativeArray x,
                      java.lang.Object y)
        Constructor which takes as arguments the x and y data. If the real response variable y is unknown you should pass a null value.
        Parameters:
        x -
        y -
      • Record

        public Record(AssociativeArray x,
                      java.lang.Object y,
                      java.lang.Object yPredicted,
                      AssociativeArray yPredictedProbabilities)
        Constructor which takes as arguments the x, y, predicted y and predicted probaibilites.
        Parameters:
        x -
        y -
        yPredicted -
        yPredictedProbabilities -
    • Method Detail

      • getX

        public AssociativeArray getX()
        It returns an AssociativeArray with all the xData of the Record. The data are stored in an unmodifiable map to ensure they can't be changed.
        Returns:
      • getY

        public java.lang.Object getY()
        It returns the real response variable of the Record. If unknown it will have a null value.
        Returns:
      • getYPredicted

        public java.lang.Object getYPredicted()
        It returns the predicted response variable of the Record. The predicted variable has a non-null value when a prediction is made by the machine learning model.
        Returns:
      • getYPredictedProbabilities

        public AssociativeArray getYPredictedProbabilities()
        If prediction probabilities are available for the particular algorithm, then this method will return these probabilities. This is supported by all the classification and some clustering algorithms.
        Returns:
      • hashCode

        public int hashCode()
        The hash code of the record. Depends only on x and y.
        Overrides:
        hashCode in class java.lang.Object
        Returns:
      • equals

        public boolean equals(java.lang.Object obj)
        Checks if the provided object is equal to the current instance. It checks only x and y.
        Overrides:
        equals in class java.lang.Object
        Parameters:
        obj -
        Returns:
      • toString

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

DataMelt 3.0 © DataMelt by jWork.ORG

You see the box below because you did not login.