/*
* RapidMiner
*
* Copyright (C) 2001-2011 by Rapid-I and the contributors
*
* Complete list of developers available at our web site:
*
* http://rapid-i.com
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program 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 Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see http://www.gnu.org/licenses/.
*/
package com.rapidminer.operator.features.transformation;
import java.util.ArrayList;
import java.util.List;
import Jama.EigenvalueDecomposition;
import Jama.Matrix;
import com.rapidminer.example.Attribute;
import com.rapidminer.example.Attributes;
import com.rapidminer.example.Example;
import com.rapidminer.example.ExampleSet;
import com.rapidminer.example.Statistics;
import com.rapidminer.example.Tools;
import com.rapidminer.operator.Model;
import com.rapidminer.operator.Operator;
import com.rapidminer.operator.OperatorDescription;
import com.rapidminer.operator.OperatorException;
import com.rapidminer.operator.ports.InputPort;
import com.rapidminer.operator.ports.OutputPort;
import com.rapidminer.operator.ports.metadata.AttributeMetaData;
import com.rapidminer.operator.ports.metadata.ExampleSetMetaData;
import com.rapidminer.operator.ports.metadata.ExampleSetPassThroughRule;
import com.rapidminer.operator.ports.metadata.ExampleSetPrecondition;
import com.rapidminer.operator.ports.metadata.PassThroughRule;
import com.rapidminer.operator.ports.metadata.SetRelation;
import com.rapidminer.parameter.ParameterType;
import com.rapidminer.parameter.UndefinedParameterError;
import com.rapidminer.tools.Ontology;
import com.rapidminer.tools.math.kernels.Kernel;
/**
* This operator performs a kernel-based principal components analysis (PCA).
* Hence, the result will be the set of data points in a non-linearly
* transformed space. Please note that in contrast to the usual linear PCA
* the kernel variant does also works for large numbers of attributes but
* will become slow for large number of examples.
*
* @author Sebastian Land
*/
public class KernelPCA extends Operator {
private InputPort exampleSetInput = getInputPorts().createPort("example set input");
private OutputPort exampleSetOutput = getOutputPorts().createPort("example set output");
private OutputPort originalOutput = getOutputPorts().createPort("original");
private OutputPort modelOutput = getOutputPorts().createPort("preprocessing model");
public KernelPCA(OperatorDescription description) {
super(description);
exampleSetInput.addPrecondition(new ExampleSetPrecondition(exampleSetInput, Ontology.NUMERICAL));
getTransformer().addRule(new ExampleSetPassThroughRule(exampleSetInput, exampleSetOutput, SetRelation.EQUAL) {
@Override
public ExampleSetMetaData modifyExampleSet(ExampleSetMetaData metaData) throws UndefinedParameterError {
switch (metaData.getNumberOfExamples().getRelation()) {
case EQUAL:
metaData.attributesAreKnown();
break;
case AT_LEAST:
metaData.attributesAreSubset();
break;
case AT_MOST:
case UNKNOWN:
metaData.attributesAreSuperset();
break;
}
if (metaData.getNumberOfExamples().getNumber() != null) {
int numberOfExamples = metaData.getNumberOfExamples().getNumber();
metaData.clearRegular();
for (int i = 1; i <= numberOfExamples; i++) {
metaData.addAttribute(new AttributeMetaData("kpc_" + i, Ontology.REAL));
}
}
return super.modifyExampleSet(metaData);
}
});
getTransformer().addRule(new PassThroughRule(exampleSetInput, originalOutput, false));
getTransformer().addGenerationRule(modelOutput, Model.class);
}
@Override
public void doWork() throws OperatorException {
ExampleSet exampleSet = exampleSetInput.getData();
// only use numeric attributes
Tools.onlyNumericalAttributes(exampleSet, "KernelPCA");
Tools.onlyNonMissingValues(exampleSet, "KernelPCA");
Attributes attributes = exampleSet.getAttributes();
int numberOfExamples = exampleSet.size();
// calculating means for later zero centering
exampleSet.recalculateAllAttributeStatistics();
double[] means = new double[exampleSet.getAttributes().size()];
int i = 0;
for (Attribute attribute: exampleSet.getAttributes()) {
means[i] = exampleSet.getStatistics(attribute, Statistics.AVERAGE);
i++;
}
// kernel
Kernel kernel = Kernel.createKernel(this);
// copying zero centered exampleValues
ArrayList<double[]> exampleValues = new ArrayList<double[]>(numberOfExamples);
i = 0;
for(Example columnExample: exampleSet) {
double[] columnValues = getAttributeValues(columnExample, attributes, means);
exampleValues.add(columnValues);
i++;
}
// filling kernel matrix
Matrix kernelMatrix = new Matrix(numberOfExamples, numberOfExamples);
for (i = 0; i < numberOfExamples; i++)
for (int j = 0; j < numberOfExamples; j++)
kernelMatrix.set(i, j, kernel.calculateDistance(exampleValues.get(i), exampleValues.get(j)));
// calculating eigenVectors
EigenvalueDecomposition eig = kernelMatrix.eig();
Model model = new KernelPCAModel(exampleSet, means, eig.getV(), exampleValues, kernel);
if (exampleSetOutput.isConnected())
exampleSetOutput.deliver(model.apply((ExampleSet)exampleSet.clone()));
originalOutput.deliver(exampleSet);
modelOutput.deliver(model);
}
private double[] getAttributeValues(Example example, Attributes attributes, double[] means) {
double[] values = new double[attributes.size()];
int x = 0;
for (Attribute attribute : attributes) {
values[x] = example.getValue(attribute) - means[x];
x++;
}
return values;
}
@Override
public List<ParameterType> getParameterTypes() {
List<ParameterType> types = super.getParameterTypes();
types.addAll(Kernel.getParameters(this));
return types;
}
}