net.sf.myra.datamining.util
Class HierarchicalCrossValidation

java.lang.Object
  extended by net.sf.myra.datamining.util.HierarchicalCrossValidation

public class HierarchicalCrossValidation
extends java.lang.Object

This class represents the cross-validation procedure for hierarchical classification problems.

Author:
Fernando Esteban Barril Otero

Constructor Summary
HierarchicalCrossValidation()
           
 
Method Summary
static void export(Dataset dataset, java.io.File directory, Helper helper, int folds)
          Exports the dataset into training, validation and test set.
static void export(Dataset dataset, java.io.File directory, Helper helper, int folds, boolean validation)
          Exports the dataset into folds partitions.
static void main(java.lang.String[] args)
          CrossValidation entry point.
static FlatCrossValidation.Partition[] partition(Dataset dataset, int folds)
          Splits the dataset into folds stratified partitions.
static Dataset[] split(Dataset dataset, int folds)
          Splits the dataset into folds stratified partitions.
static Dataset[] split(java.lang.String filename, int folds)
          Splits the dataset into folds stratified partitions.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

HierarchicalCrossValidation

public HierarchicalCrossValidation()
Method Detail

partition

public static FlatCrossValidation.Partition[] partition(Dataset dataset,
                                                        int folds)
Splits the dataset into folds stratified partitions.

Parameters:
dataset - the dataset to split.
folds - the number of cross-validation folds.
Returns:
an array containing folds stratified partitions.

split

public static Dataset[] split(java.lang.String filename,
                              int folds)
                       throws java.io.IOException
Splits the dataset into folds stratified partitions.

Parameters:
filename - the dataset filename to split.
folds - the number of cross-validation folds.
Returns:
an array containing folds stratified Dataset instances.
Throws:
java.io.IOException
See Also:
Dataset

split

public static Dataset[] split(Dataset dataset,
                              int folds)
                       throws java.io.IOException
Splits the dataset into folds stratified partitions.

Parameters:
dataset - the dataset to split.
folds - the number of cross-validation folds.
Returns:
an array containing folds stratified Dataset instances.
Throws:
java.io.IOException
See Also:
Dataset

export

public static void export(Dataset dataset,
                          java.io.File directory,
                          Helper helper,
                          int folds,
                          boolean validation)
                   throws java.io.IOException
Exports the dataset into folds partitions. For each partiton, a training and test file is generated.

Parameters:
dataset - the dataset to export.
directory - the target directory.
helper - the file helper.
folds - the number of folds.
Throws:
java.io.IOException - If an I/O error has occurred.

export

public static void export(Dataset dataset,
                          java.io.File directory,
                          Helper helper,
                          int folds)
                   throws java.io.IOException
Exports the dataset into training, validation and test set.

Parameters:
dataset - the dataset to export.
directory - the target directory.
helper - the file helper.
folds - the number of folds.
Throws:
java.io.IOException - If an I/O error has occurred.

main

public static void main(java.lang.String[] args)
                 throws java.lang.Exception
CrossValidation entry point.

Parameters:
args - an array of command line arguments.
Throws:
java.lang.Exception


Copyright © 2013. All Rights Reserved.