net.sourceforge.cilib.problem.dataset
Class LocalDataSet

java.lang.Object
  net.sourceforge.cilib.problem.dataset.DataSet
      net.sourceforge.cilib.problem.dataset.LocalDataSet

All Implemented Interfaces:

Serializable, Cloneable

public class LocalDataSet
extends DataSet

This class represents a local dataset, i.e. a local file on disk that contains lines that represent patterns of the dataset. It is responsible for parsing this file and building up an ArrayList of ClusterableDataSet.Pattern objects. It makes use of a few variables to correctly parse the lines in the dataset and construct ClusterableDataSet.Pattern objects from them. The first variable that is of importance is the delimiter variable. It specifies the delimiter (actually a regular expression) that is used to split up the elements of a single line of the dataset. The default delimiter is a whitespace character. Once that is done, the beginIndex specifies the column number/index where the pattern's data begins. Likewise, the endIndex specifies the column number/index where the pattern's data ends. This index is inclusive. Lastly, the classIndex specifies the column number/index that represents the pattern's class. If the classIndex is -1 then it means that the dataset does not have a column for the class of the patterns, and in this case the filename of the dataset is used as the class.

Author:

Edwin Peer, Theuns Cloete

See Also:

Serialized Form

Field Summary

protected int	beginIndex
protected int	classIndex
protected String	delimiter
protected int	endIndex
protected String	fileName

Fields inherited from class net.sourceforge.cilib.problem.dataset.DataSet

patternExpression

Constructor Summary

LocalDataSet()

LocalDataSet(LocalDataSet rhs)

Method Summary

LocalDataSet	getClone() Create a cloned copy of the current object and return it.
byte[]	getData() Get the contents of the file on disk as an array of bytes.
String	getFile() Get the name of the file that represents this dataset on disk.
InputStream	getInputStream() Get the contents of the file on disk as an InputStream.
ArrayList<ClusterableDataSet.Pattern>	parseDataSet() Parse the dataset, building up a list containing all the patterns in the dataset.
void	setBeginIndex(int bi) Sets the index where the elements of the pattern begins.
void	setClassIndex(int ci) Sets the index of the column that represents the class of the pattern.
void	setDelimiter(String d) Sets the regular expression (as a String} that should be used as delimiter to split a string into the elements of the pattern.
void	setEndIndex(int ei) Sets the index where the elements of the pattern ends.
void	setFile(String fileName) Set the name of the file that represents this dataset on disk.

Methods inherited from class net.sourceforge.cilib.problem.dataset.DataSet

getPatternExpression, setPatternExpression

Methods inherited from class java.lang.Object

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

Field Detail

fileName

protected String fileName

delimiter

protected String delimiter

beginIndex

protected int beginIndex

endIndex

protected int endIndex

classIndex

protected int classIndex

Constructor Detail

LocalDataSet

public LocalDataSet()

LocalDataSet

public LocalDataSet(LocalDataSet rhs)

Method Detail

getClone

public LocalDataSet getClone()

Description copied from interface: Cloneable

Create a cloned copy of the current object and return it. In general the created copy will be a deep copy of the provided instance. As a result this operation an be quite expensive if used incorrectly.

Specified by:

getClone in interface Cloneable

Specified by:

getClone in class DataSet

Returns:

An exact clone of the current object instance.

See Also:

Object.clone()

setFile

public void setFile(String fileName)

Set the name of the file that represents this dataset on disk.

Parameters:

fileName - the name of the file

getFile

public String getFile()

Get the name of the file that represents this dataset on disk.

Returns:

the name of the file

getData

public byte[] getData()

Get the contents of the file on disk as an array of bytes.

Specified by:

getData in class DataSet

Returns:

the contents of the file on disk as an array of bytes

getInputStream

public InputStream getInputStream()

Get the contents of the file on disk as an InputStream.

Specified by:

getInputStream in class DataSet

Returns:

the contents of the file on disk as an InputStream

parseDataSet

public ArrayList<ClusterableDataSet.Pattern> parseDataSet()

Parse the dataset, building up a list containing all the patterns in the dataset.

Returns:

an ArrayList of ClusterableDataSet.Patterns containing all the patterns in this dataset

Throws:

IllegalArgumentException - when beginIndex == endIndex.

setDelimiter

public void setDelimiter(String d)

Sets the regular expression (as a String} that should be used as delimiter to split a string into the elements of the pattern.

Parameters:

d - the regular expression (as a String) that should be used as delimiter

Throws:

IllegalArgumentException - when the delimiter is empty ("") or null

setBeginIndex

public void setBeginIndex(int bi)

Sets the index where the elements of the pattern begins.

Parameters:

bi - the starting index

Throws:

IllegalArgumentException - when the index is negative

setEndIndex

public void setEndIndex(int ei)

Sets the index where the elements of the pattern ends. This index is inclusive.

Parameters:

ei - the ending index

Throws:

IllegalArgumentException - when the index is negative

setClassIndex

public void setClassIndex(int ci)

Sets the index of the column that represents the class of the pattern. If the index is -1 then the filename of the dataset will be used as the class.

Parameters:

ci - the index where the class resides

Throws:

IllegalArgumentException - when the index is <-1