jsat.classifiers

Class DataPoint

java.lang.Object

jsat.classifiers.DataPoint

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable

public class DataPoint
extends java.lang.Object
implements java.lang.Cloneable, java.io.Serializable

This is the general class object for representing a singular data point in a data set. Every data point is made up of either categorical variables, numerical variables, or a combination of the two.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
DataPoint(Vec numericalValues) Creates a new data point that has no categorical variables and a weight of 1.0
DataPoint(Vec numericalValues, double weight) Creates a new data point that has no categorical variables
DataPoint(Vec numericalValues, int[] categoricalValues, CategoricalData[] categoricalData) Creates a new data point with the default weight of 1.0
DataPoint(Vec numericalValues, int[] categoricalValues, CategoricalData[] categoricalData, double weight) Creates a new data point

Method Summary

Modifier and Type	Method and Description
DataPoint	clone() Creates a deep clone of this data point, such that altering either data point does not effect the other one.
boolean	containsCategoricalData() Returns true if this data point contains any categorical variables, false otherwise.
boolean	containsNumericalData() Returns true if the data point contains any numerical variables, false otherwise.
CategoricalData[]	getCategoricalData() Returns the array of Categorical Data information
int	getCategoricalValue(int i)
int[]	getCategoricalValues() Returns the array of values for each category.
Vec	getNumericalValues() Returns the vector containing the numerical values.
double	getWeight() Returns the weight that this data point carries.
int	numCategoricalValues() Returns the number of categorical variables in this data point.
int	numNumericalValues() Returns the number of numerical variables in this data point.
void	setWeight(double weight) Set the weight that this data point should carry.
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

DataPoint

public DataPoint(Vec numericalValues,
                 int[] categoricalValues,
                 CategoricalData[] categoricalData)

Creates a new data point with the default weight of 1.0

Parameters:

numericalValues - a vector containing the numerical values for this data point

categoricalValues - an array of the category values for this data point

categoricalData - an array of the category information of this data point

DataPoint

public DataPoint(Vec numericalValues,
                 int[] categoricalValues,
                 CategoricalData[] categoricalData,
                 double weight)

Creates a new data point

Parameters:

numericalValues - a vector containing the numerical values for this data point

categoricalValues - an array of the category values for this data point

categoricalData - an array of the category information of this data point

weight - a double indicating how much weight this data point has, 1.0 being the standard weight. Not all algorithms make use of data points that have different weights.

DataPoint

public DataPoint(Vec numericalValues,
                 double weight)

Creates a new data point that has no categorical variables

Parameters:

numericalValues - a vector containing the numerical values for this data point

weight - a double indicating how much weight this data point has, 1.0 being the standard weight. Not all algorithms make use of data points that have different weights.

DataPoint

public DataPoint(Vec numericalValues)

Creates a new data point that has no categorical variables and a weight of 1.0

Parameters:

numericalValues - a vector containing the numerical values for this data point

Method Detail

getWeight

public double getWeight()

Returns the weight that this data point carries.

Returns:

the weight that this data point carries.

setWeight

public void setWeight(double weight)

Set the weight that this data point should carry. The norm is 1.0

Parameters:

weight - the new weight value

Throws:

java.lang.ArithmeticException - if the weight value is not a number > 0

containsCategoricalData

public boolean containsCategoricalData()

Returns true if this data point contains any categorical variables, false otherwise.

Returns:

true if this data point contains any categorical variables, false otherwise.

getNumericalValues

public Vec getNumericalValues()

Returns the vector containing the numerical values. Altering this vector will effect this data point. If changes are going to be made, a clone of the vector should be made by the caller.

Returns:

the vector containing the numerical values.

containsNumericalData

public boolean containsNumericalData()

Returns true if the data point contains any numerical variables, false otherwise.

Returns:

true if the data point contains any numerical variables, false otherwise.

numNumericalValues

public int numNumericalValues()

Returns the number of numerical variables in this data point.

Returns:

the number of numerical variables in this data point.

getCategoricalValues

public int[] getCategoricalValues()

Returns the array of values for each category. Altering this array will effect this data point. If changes are going to be made, a clone of the array should be made by the caller.

Returns:

the array of values for each category.

numCategoricalValues

public int numCategoricalValues()

Returns the number of categorical variables in this data point.

Returns:

the number of categorical variables in this data point.

getCategoricalValue

public int getCategoricalValue(int i)

Parameters:

i - the i'th categorical variable

Returns:

the value of the i'th categorical variable

getCategoricalData

public CategoricalData[] getCategoricalData()

Returns the array of Categorical Data information

Returns:

the array of Categorical Data information

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

clone

public DataPoint clone()

Creates a deep clone of this data point, such that altering either data point does not effect the other one.

Overrides:

clone in class java.lang.Object

Returns:

a deep clone of this data point.