package shared.test;
import shared.DataSet;
import shared.Instance;
import shared.filt.PrincipalComponentAnalysis;
import util.linalg.Matrix;
/**
* A class for testing
* @author Andrew Guillory gtg008g@mail.gatech.edu
* @version 1.0
*/
public class PrincipalComponentAnalysisTest {
/**
* The test main
* @param args ignored
*/
public static void main(String[] args) {
Instance[] instances = {
new Instance(new double[] {1,1,0,0,0,0,0,0}),
new Instance(new double[] {0,0,1,1,1,0,0,0}),
new Instance(new double[] {0,0,0,0,1,1,1,1}),
new Instance(new double[] {1,0,1,0,1,0,1,0}),
new Instance(new double[] {1,1,0,0,1,1,0,0}),
};
DataSet set = new DataSet(instances);
System.out.println("Before PCA");
System.out.println(set);
PrincipalComponentAnalysis filter = new PrincipalComponentAnalysis(set);
System.out.println(filter.getEigenValues());
System.out.println(filter.getProjection().transpose());
filter.filter(set);
System.out.println("After PCA");
System.out.println(set);
Matrix reverse = filter.getProjection().transpose();
for (int i = 0; i < set.size(); i++) {
Instance instance = set.get(i);
instance.setData(reverse.times(instance.getData()).plus(filter.getMean()));
}
System.out.println("After reconstructing");
System.out.println(set);
}
}