// Catalano Machine Learning Library
// The Catalano Framework
//
// Copyright © Diego Catalano, 2012-2018
// diego.catalano at live.com
//
//
// This library is free software; you can redistribute it and/or
// modify it under the terms of the GNU Lesser General Public
// License as published by the Free Software Foundation; either
// version 2.1 of the License, or (at your option) any later version.
//
// This library is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
// Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public
// License along with this library; if not, write to the Free Software
// Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
//
package Catalano.MachineLearning.Performance;
import Catalano.Core.ArraysUtil;
import Catalano.MachineLearning.Classification.IClassifier;
import Catalano.MachineLearning.Dataset.DatasetClassification;
import Catalano.Math.Matrix;
/**
* KFold Cross Validation.
* @author Diego Catalano
*/
public class KFoldCrossValidation implements IValidation{
private int nFolds;
private boolean shuffle;
private long seed;
/**
* Get number of the folds.
* @return Number of the folds.
*/
public int getNumberOfFolds(){
return nFolds;
}
/**
* Set number of the folds.
* @param folds Folds.
*/
public void setNumberOfFolds(int folds){
Math.max(folds, 2);
}
/**
* Checks if the indexes are going to be shuffled.
* @return True if the indexes are going to be shuffled, othwerwise return false.
*/
public boolean isShuffle() {
return shuffle;
}
/**
* Set true if the indexes are going to be shuffled.
* @param shuffle Indexes are going to be shuffled.
*/
public void setShuffle(boolean shuffle) {
this.shuffle = shuffle;
}
/**
* Initializes a new instance of the KFoldCrossValidation class.
*/
public KFoldCrossValidation() {
this(10);
}
/**
* Initializes a new instance of the KFoldCrossValidation class.
* @param nFolds Number of folds.
*/
public KFoldCrossValidation(int nFolds) {
this(nFolds, true);
}
/**
* Initializes a new instance of the KFoldCrossValidation class.
* @param nFolds Number of folds.
* @param shuffle Shuffle the indexes.
*/
public KFoldCrossValidation(int nFolds, boolean shuffle) {
this(nFolds, shuffle, -1);
}
/**
* Initializes a new instance of the KFoldCrossValidation class.
* @param nFolds Number of folds.
* @param shuffle Shuffle the indexes.
* @param seed Random seed generator.
*/
public KFoldCrossValidation(int nFolds, boolean shuffle, long seed){
this.nFolds = nFolds;
this.shuffle = shuffle;
this.seed = seed;
}
@Override
public double Run(IClassifier classifier, DatasetClassification dataset) {
return Run(classifier, dataset.getInput(), dataset.getOutput());
}
@Override
public double Run(IClassifier classifier, double[][] data, int[] labels) {
if(nFolds > data.length)
throw new IllegalArgumentException("The number of folds must be less or equal than number of samples.");
int parts;
double p = (labels.length / (double)nFolds) - labels.length / nFolds;
if(p > 0.5)
parts = (labels.length / nFolds) + 1;
else
parts = (labels.length / nFolds);
int start = 0;
int end = 0;
int[] indexes = Matrix.Indices(0, labels.length);
if(shuffle && seed == 0){
ArraysUtil.Shuffle(indexes);
}
else if(shuffle && seed != 0){
ArraysUtil.Shuffle(indexes, seed);
}
double mean = 0;
for (int i = 0; i < nFolds; i++) {
//The last fold, we need to select all the rest.
if(i == nFolds - 1){
end = indexes.length;
}
else{
end += parts;
}
//Get the indexes for train/test
int[] ind = Matrix.getColumns(indexes, Matrix.Indices(start, end));
//Train data
double[][] inputTrain = Matrix.RemoveRows(data, ind);
int[] outputTrain = Matrix.RemoveColumns(labels, ind);
//Test data
double[][] inputTest = Matrix.getRows(data, Matrix.Indices(start, end));
int[] outputTest = Matrix.getRows(labels, Matrix.Indices(start, end));
classifier.Learn(inputTrain, outputTrain);
SuppliedValidation sv = new SuppliedValidation();
mean += sv.Run(classifier, inputTest, outputTest);
start = end;
}
return mean/nFolds;
}
}
Ads help maintain this website.