edu.columbia.cs.ref.tool.collection.splitter
Class Splitter<E extends Writable>

java.lang.Object
  edu.columbia.cs.ref.tool.collection.splitter.Splitter<E>

Direct Known Subclasses:

KFoldSplitter

public abstract class Splitter<E extends Writable>
extends java.lang.Object

Interface for a Splitter

A splitter is a utility that is not directly used in a relationship extraction process but that is very useful for evaluation purposes.

It is expected that a class that implements splitter separates a dataset into several folds in order for it to be used on cross-validation

Since:

2011-09-27

Version:

0.1

Author:

Pablo Barrio, Goncalo Simoes

Constructor Summary

Splitter()

Method Summary

abstract void

split(Dataset<E> directory, java.io.File outputFolder) Given a dataset, this method is responsible for creating the folds for it.

Methods inherited from class java.lang.Object

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

Constructor Detail

Splitter

public Splitter()

Method Detail

split

public abstract void split(Dataset<E> directory,
                           java.io.File outputFolder)

Given a dataset, this method is responsible for creating the folds for it. The results are written to the folder given as input

Parameters:

dataset - dataset to be splitted

outputFolder - folder where the fold files will be written