weka.core

Interface Instance

All Superinterfaces:

Copyable

All Known Implementing Classes:

AbstractInstance, BinarySparseInstance, DenseInstance, SparseInstance

public interface Instance
extends Copyable

Interface representing an instance. All values (numeric, date, nominal, string or relational) are internally stored as floating-point numbers in the original concrete class implementations (now called DenseInstance.java and SparseInstance.java), and the methods in this interface reflect this. If an attribute is nominal (or a string or relational), the stored value is the index of the corresponding nominal (or string or relational) value in the attribute's definition. We have chosen this approach in favor of a more elegant object-oriented approach because it is much faster.

Typical usage (code from the main() method of this class):

...
// Create empty instance with three attribute values
Instance inst = new DenseInstance(3);

// Set instance's values for the attributes "length", "weight", and "position"
inst.setValue(length, 5.3);
inst.setValue(weight, 300);
inst.setValue(position, "first");

// Set instance's dataset to be the dataset "race"
inst.setDataset(race);

// Print the instance
System.out.println("The instance: " + inst);
...


All methods that change an instance's attribute values must be safe, ie. a change of an instance's attribute values must not affect any other instances.

Version:

$Revision: 15070 $

Author:

Eibe Frank (eibe@cs.waikato.ac.nz)

Method Summary

Modifier and Type	Method and Description
Attribute	attribute(int index) Returns the attribute with the given index.
Attribute	attributeSparse(int indexOfIndex) Returns the attribute with the given index in the sparse representation.
Attribute	classAttribute() Returns class attribute.
int	classIndex() Returns the class attribute's index.
boolean	classIsMissing() Tests if an instance's class is missing.
double	classValue() Returns an instance's class value as a floating-point number.
Instance	copy(double[] values) Copies the instance but fills up its values based on the given array of doubles.
Instances	dataset() Returns the dataset this instance has access to.
void	deleteAttributeAt(int position) Deletes an attribute at the given position (0 to numAttributes() - 1).
java.util.Enumeration<Attribute>	enumerateAttributes() Returns an enumeration of all the attributes.
boolean	equalHeaders(Instance inst) Tests if the headers of two instances are equivalent.
java.lang.String	equalHeadersMsg(Instance inst) Checks if the headers of two instances are equivalent.
boolean	hasMissingValue() Tests whether an instance has a missing value.
int	index(int position) Returns the index of the attribute stored at the given position in the sparse representation.
void	insertAttributeAt(int position) Inserts an attribute at the given position (0 to numAttributes()).
boolean	isMissing(Attribute att) Tests if a specific value is "missing".
boolean	isMissing(int attIndex) Tests if a specific value is "missing".
boolean	isMissingSparse(int indexOfIndex) Tests if a specific value is "missing" in the sparse representation.
Instance	mergeInstance(Instance inst) Merges this instance with the given instance and returns the result.
int	numAttributes() Returns the number of attributes.
int	numClasses() Returns the number of class labels.
int	numValues() Returns the number of values present in a sparse representation.
Instances	relationalValue(Attribute att) Returns the relational value of a relational attribute.
Instances	relationalValue(int attIndex) Returns the relational value of a relational attribute.
void	replaceMissingValues(double[] array) Replaces all missing values in the instance with the values contained in the given array.
void	setClassMissing() Sets the class value of an instance to be "missing".
void	setClassValue(double value) Sets the class value of an instance to the given value (internal floating-point format).
void	setClassValue(java.lang.String value) Sets the class value of an instance to the given value.
void	setDataset(Instances instances) Sets the reference to the dataset.
void	setMissing(Attribute att) Sets a specific value to be "missing".
void	setMissing(int attIndex) Sets a specific value to be "missing".
void	setValue(Attribute att, double value) Sets a specific value in the instance to the given value (internal floating-point format).
void	setValue(Attribute att, java.lang.String value) Sets a value of an nominal or string attribute to the given value.
void	setValue(int attIndex, double value) Sets a specific value in the instance to the given value (internal floating-point format).
void	setValue(int attIndex, java.lang.String value) Sets a value of a nominal or string attribute to the given value.
void	setValueSparse(int indexOfIndex, double value) Sets a specific value in the instance to the given value (internal floating-point format), given an index into the sparse representation.
void	setWeight(double weight) Sets the weight of an instance.
java.lang.String	stringValue(Attribute att) Returns the value of a nominal, string, date, or relational attribute for the instance as a string.
java.lang.String	stringValue(int attIndex) Returns the value of a nominal, string, date, or relational attribute for the instance as a string.
double[]	toDoubleArray() Returns the values of each attribute as an array of doubles.
java.lang.String	toString(Attribute att) Returns the description of one value of the instance as a string.
java.lang.String	toString(Attribute att, int afterDecimalPoint) Returns the description of one value of the instance as a string.
java.lang.String	toString(int attIndex) Returns the description of one value of the instance as a string.
java.lang.String	toString(int attIndex, int afterDecimalPoint) Returns the description of one value of the instance as a string.
java.lang.String	toStringMaxDecimalDigits(int afterDecimalPoint) Returns the description of one instance with any numeric values printed at the supplied maximum number of decimal places.
java.lang.String	toStringNoWeight() Returns the description of one instance (without weight appended).
java.lang.String	toStringNoWeight(int afterDecimalPoint) Returns the description of one instance (without weight appended).
double	value(Attribute att) Returns an instance's attribute value in internal format.
double	value(int attIndex) Returns an instance's attribute value in internal format.
double	valueSparse(int indexOfIndex) Returns an instance's attribute value in internal format, given an index in the sparse representation.
double	weight() Returns the instance's weight.

Methods inherited from interface weka.core.Copyable

copy

Method Detail

attribute

Attribute attribute(int index)

Returns the attribute with the given index.

Parameters:

index - the attribute's index

Returns:

the attribute at the given position

Throws:

UnassignedDatasetException - if instance doesn't have access to a dataset

attributeSparse

Attribute attributeSparse(int indexOfIndex)

Returns the attribute with the given index in the sparse representation. Same as attribute(int) for a DenseInstance.

Parameters:

indexOfIndex - the index of the attribute's index

Returns:

the attribute at the given position

Throws:

UnassignedDatasetException - if instance doesn't have access to a dataset

classAttribute

Attribute classAttribute()

Returns class attribute.

Returns:

the class attribute

Throws:

UnassignedDatasetException - if the class is not set or the instance doesn't have access to a dataset

classIndex

int classIndex()

Returns the class attribute's index.

Returns:

the class index as an integer

Throws:

UnassignedDatasetException - if instance doesn't have access to a dataset

classIsMissing

boolean classIsMissing()

Tests if an instance's class is missing.

Returns:

true if the instance's class is missing

Throws:

UnassignedClassException - if the class is not set or the instance doesn't have access to a dataset

classValue

double classValue()

Returns an instance's class value as a floating-point number.

Returns:

the corresponding value as a double (If the corresponding attribute is nominal (or a string) then it returns the value's index as a double).

Throws:

UnassignedClassException - if the class is not set or the instance doesn't have access to a dataset

copy

Instance copy(double[] values)

Copies the instance but fills up its values based on the given array of doubles. The copy has access to the same dataset.

Parameters:

values - the array with new values

Returns:

the new instance

dataset

Instances dataset()

Returns the dataset this instance has access to. (ie. obtains information about attribute types from) Null if the instance doesn't have access to a dataset.

Returns:

the dataset the instance has accesss to

deleteAttributeAt

void deleteAttributeAt(int position)

Deletes an attribute at the given position (0 to numAttributes() - 1). Only succeeds if the instance does not have access to any dataset because otherwise inconsistencies could be introduced.

Parameters:

position - the attribute's position

Throws:

java.lang.RuntimeException - if the instance has access to a dataset

enumerateAttributes

java.util.Enumeration<Attribute> enumerateAttributes()

Returns an enumeration of all the attributes.

Returns:

enumeration of all the attributes

Throws:

UnassignedDatasetException - if the instance doesn't have access to a dataset

equalHeaders

boolean equalHeaders(Instance inst)

Tests if the headers of two instances are equivalent.

Parameters:

inst - another instance

Returns:

true if the header of the given instance is equivalent to this instance's header

Throws:

UnassignedDatasetException - if instance doesn't have access to any dataset

equalHeadersMsg

java.lang.String equalHeadersMsg(Instance inst)

Checks if the headers of two instances are equivalent. If not, then returns a message why they differ.

Parameters:

inst - another instance

Returns:

null if the header of the given instance is equivalent to this instance's header, otherwise a message with details on why they differ

hasMissingValue

boolean hasMissingValue()

Tests whether an instance has a missing value. Skips the class attribute if set.

Returns:

true if instance has a missing value.

Throws:

UnassignedDatasetException - if instance doesn't have access to any dataset

index

int index(int position)

Returns the index of the attribute stored at the given position in the sparse representation. Identify function for an instance of type DenseInstance.

Parameters:

position - the position

Returns:

the index of the attribute stored at the given position

insertAttributeAt

void insertAttributeAt(int position)

Inserts an attribute at the given position (0 to numAttributes()). Only succeeds if the instance does not have access to any dataset because otherwise inconsistencies could be introduced.

Parameters:

position - the attribute's position

Throws:

java.lang.RuntimeException - if the instance has accesss to a dataset

java.lang.IllegalArgumentException - if the position is out of range

isMissing

boolean isMissing(int attIndex)

Tests if a specific value is "missing".

Parameters:

attIndex - the attribute's index

Returns:

true if the value is "missing"

isMissingSparse

boolean isMissingSparse(int indexOfIndex)

Tests if a specific value is "missing" in the sparse representation. Samse as isMissing(int) for a DenseInstance.

Parameters:

indexOfIndex - the index of the attribute's index

Returns:

true if the value is "missing"

isMissing

boolean isMissing(Attribute att)

Tests if a specific value is "missing". The given attribute has to belong to a dataset.

Parameters:

att - the attribute

Returns:

true if the value is "missing"

mergeInstance

Instance mergeInstance(Instance inst)

Merges this instance with the given instance and returns the result. Dataset is set to null. The returned instance is of the same type as this instance.

Parameters:

inst - the instance to be merged with this one

Returns:

the merged instances

numAttributes

int numAttributes()

Returns the number of attributes.

Returns:

the number of attributes as an integer

numClasses

int numClasses()

Returns the number of class labels.

Returns:

the number of class labels as an integer if the class attribute is nominal, 1 otherwise.

Throws:

UnassignedDatasetException - if instance doesn't have access to any dataset

numValues

int numValues()

Returns the number of values present in a sparse representation.

Returns:

the number of values

replaceMissingValues

void replaceMissingValues(double[] array)

Replaces all missing values in the instance with the values contained in the given array. A deep copy of the vector of attribute values is performed before the values are replaced.

Parameters:

array - containing the means and modes

Throws:

java.lang.IllegalArgumentException - if numbers of attributes are unequal

setClassMissing

void setClassMissing()

Sets the class value of an instance to be "missing". A deep copy of the vector of attribute values is performed before the value is set to be missing.

Throws:

UnassignedClassException - if the class is not set

UnassignedDatasetException - if the instance doesn't have access to a dataset

setClassValue

void setClassValue(double value)

Sets the class value of an instance to the given value (internal floating-point format). A deep copy of the vector of attribute values is performed before the value is set.

Parameters:

value - the new attribute value (If the corresponding attribute is nominal (or a string) then this is the new value's index as a double).

Throws:

UnassignedClassException - if the class is not set

UnassignedDatasetException - if the instance doesn't have access to a dataset

setClassValue

void setClassValue(java.lang.String value)

Sets the class value of an instance to the given value. A deep copy of the vector of attribute values is performed before the value is set.

Parameters:

value - the new class value (If the class is a string attribute and the value can't be found, the value is added to the attribute).

Throws:

UnassignedClassException - if the class is not set

UnassignedDatasetException - if the dataset is not set

java.lang.IllegalArgumentException - if the attribute is not nominal or a string, or the value couldn't be found for a nominal attribute

setDataset

void setDataset(Instances instances)

Sets the reference to the dataset. Does not check if the instance is compatible with the dataset. Note: the dataset does not know about this instance. If the structure of the dataset's header gets changed, this instance will not be adjusted automatically.

Parameters:

instances - the reference to the dataset

setMissing

void setMissing(int attIndex)

Sets a specific value to be "missing". Performs a deep copy of the vector of attribute values before the value is set to be missing.

Parameters:

attIndex - the attribute's index

setMissing

void setMissing(Attribute att)

Sets a specific value to be "missing". Performs a deep copy of the vector of attribute values before the value is set to be missing. The given attribute has to belong to a dataset.

Parameters:

att - the attribute

setValue

void setValue(int attIndex,
              double value)

Sets a specific value in the instance to the given value (internal floating-point format). Performs a deep copy of the vector of attribute values before the value is set.

Parameters:

attIndex - the attribute's index

value - the new attribute value (If the corresponding attribute is nominal (or a string) then this is the new value's index as a double).

setValueSparse

void setValueSparse(int indexOfIndex,
                    double value)

Sets a specific value in the instance to the given value (internal floating-point format), given an index into the sparse representation. Performs a deep copy of the vector of attribute values before the value is set. Same as setValue(int, double) for a DenseInstance.

Parameters:

indexOfIndex - the index of the attribute's index

value - the new attribute value (If the corresponding attribute is nominal (or a string) then this is the new value's index as a double).

setValue

void setValue(int attIndex,
              java.lang.String value)

Sets a value of a nominal or string attribute to the given value. Performs a deep copy of the vector of attribute values before the value is set.

Parameters:

attIndex - the attribute's index

value - the new attribute value (If the attribute is a string attribute and the value can't be found, the value is added to the attribute).

Throws:

UnassignedDatasetException - if the dataset is not set

java.lang.IllegalArgumentException - if the selected attribute is not nominal or a string, or the supplied value couldn't be found for a nominal attribute

setValue

void setValue(Attribute att,
              double value)

Sets a specific value in the instance to the given value (internal floating-point format). Performs a deep copy of the vector of attribute values before the value is set, so if you are planning on calling setValue many times it may be faster to create a new instance using toDoubleArray. The given attribute has to belong to a dataset.

Parameters:

att - the attribute

value - the new attribute value (If the corresponding attribute is nominal (or a string) then this is the new value's index as a double).

setValue

void setValue(Attribute att,
              java.lang.String value)

Sets a value of an nominal or string attribute to the given value. Performs a deep copy of the vector of attribute values before the value is set, so if you are planning on calling setValue many times it may be faster to create a new instance using toDoubleArray. The given attribute has to belong to a dataset.

Parameters:

att - the attribute

value - the new attribute value (If the attribute is a string attribute and the value can't be found, the value is added to the attribute).

Throws:

java.lang.IllegalArgumentException - if the the attribute is not nominal or a string, or the value couldn't be found for a nominal attribute

setWeight

void setWeight(double weight)

Sets the weight of an instance.

Parameters:

weight - the weight

relationalValue

Instances relationalValue(int attIndex)

Returns the relational value of a relational attribute.

Parameters:

attIndex - the attribute's index

Returns:

the corresponding relation as an Instances object

Throws:

java.lang.IllegalArgumentException - if the attribute is not a relation-valued attribute

UnassignedDatasetException - if the instance doesn't belong to a dataset.

relationalValue

Instances relationalValue(Attribute att)

Returns the relational value of a relational attribute.

Parameters:

att - the attribute

Returns:

the corresponding relation as an Instances object

Throws:

java.lang.IllegalArgumentException - if the attribute is not a relation-valued attribute

UnassignedDatasetException - if the instance doesn't belong to a dataset.

stringValue

java.lang.String stringValue(int attIndex)

Returns the value of a nominal, string, date, or relational attribute for the instance as a string.

Parameters:

attIndex - the attribute's index

Returns:

the value as a string

Throws:

java.lang.IllegalArgumentException - if the attribute is not a nominal, string, date, or relation-valued attribute.

UnassignedDatasetException - if the instance doesn't belong to a dataset.

stringValue

java.lang.String stringValue(Attribute att)

Returns the value of a nominal, string, date, or relational attribute for the instance as a string.

Parameters:

att - the attribute

Returns:

the value as a string

Throws:

java.lang.IllegalArgumentException - if the attribute is not a nominal, string, date, or relation-valued attribute.

UnassignedDatasetException - if the instance doesn't belong to a dataset.

toDoubleArray

double[] toDoubleArray()

Returns the values of each attribute as an array of doubles. Creates a fresh array object for this.

Returns:

an array containing all the instance attribute values

toStringNoWeight

java.lang.String toStringNoWeight(int afterDecimalPoint)

Returns the description of one instance (without weight appended). If the instance doesn't have access to a dataset, it returns the internal floating-point values. Quotes string values that contain whitespace characters. This method is used by getRandomNumberGenerator() in Instances.java in order to maintain backwards compatibility with weka 3.4.

Parameters:

afterDecimalPoint - maximum number of digits after the decimal point for numeric values

Returns:

the instance's description as a string

toStringNoWeight

java.lang.String toStringNoWeight()

Returns the description of one instance (without weight appended). If the instance doesn't have access to a dataset, it returns the internal floating-point values. Quotes string values that contain whitespace characters. This method is used by getRandomNumberGenerator() in Instances.java in order to maintain backwards compatibility with weka 3.4.

Returns:

the instance's description as a string

toStringMaxDecimalDigits

java.lang.String toStringMaxDecimalDigits(int afterDecimalPoint)

Returns the description of one instance with any numeric values printed at the supplied maximum number of decimal places. If the instance doesn't have access to a dataset, it returns the internal floating-point values. Quotes string values that contain whitespace characters.

Parameters:

afterDecimalPoint - the maximum number of digits permitted after the decimal point for a numeric value

Returns:

the instance's description as a string

toString

java.lang.String toString(int attIndex,
                          int afterDecimalPoint)

Returns the description of one value of the instance as a string. If the instance doesn't have access to a dataset, it returns the internal floating-point value. Quotes string values that contain whitespace characters, or if they are a question mark.

Parameters:

attIndex - the attribute's index

afterDecimalPoint - the maximum number of digits permitted after the decimal point for numeric values

Returns:

the value's description as a string

toString

java.lang.String toString(int attIndex)

Returns the description of one value of the instance as a string. If the instance doesn't have access to a dataset, it returns the internal floating-point value. Quotes string values that contain whitespace characters, or if they are a question mark.

Parameters:

attIndex - the attribute's index

Returns:

the value's description as a string

toString

java.lang.String toString(Attribute att,
                          int afterDecimalPoint)

Returns the description of one value of the instance as a string. If the instance doesn't have access to a dataset it returns the internal floating-point value. Quotes string values that contain whitespace characters, or if they are a question mark. The given attribute has to belong to a dataset.

Parameters:

att - the attribute

afterDecimalPoint - the maximum number of decimal places to print

Returns:

the value's description as a string

toString

java.lang.String toString(Attribute att)

Returns the description of one value of the instance as a string. If the instance doesn't have access to a dataset it returns the internal floating-point value. Quotes string values that contain whitespace characters, or if they are a question mark. The given attribute has to belong to a dataset.

Parameters:

att - the attribute

Returns:

the value's description as a string

value

double value(int attIndex)

Returns an instance's attribute value in internal format.

Parameters:

attIndex - the attribute's index

Returns:

the specified value as a double (If the corresponding attribute is nominal (or a string) then it returns the value's index as a double).

valueSparse

double valueSparse(int indexOfIndex)

Returns an instance's attribute value in internal format, given an index in the sparse representation. Same as value(int) for a DenseInstance.

Parameters:

indexOfIndex - the index of the attribute's index

Returns:

the specified value as a double (If the corresponding attribute is nominal (or a string) then it returns the value's index as a double).

value

double value(Attribute att)

Returns an instance's attribute value in internal format. The given attribute has to belong to a dataset.

Parameters:

att - the attribute

Returns:

the specified value as a double (If the corresponding attribute is nominal (or a string) then it returns the value's index as a double).

weight

double weight()

Returns the instance's weight.

Returns:

the instance's weight as a double