CrossValidator.java

package edu.usc.cssl.tacit.classify.svm.services;

import java.io.BufferedWriter;
import java.io.File;
import java.io.FileWriter;
import java.io.IOException;
import java.text.DateFormat;
import java.text.SimpleDateFormat;
import java.util.Date;

import org.eclipse.core.runtime.IProgressMonitor;

import edu.usc.cssl.tacit.common.TacitUtility;
import edu.usc.cssl.tacit.common.ui.views.ConsoleView;

public class CrossValidator {

	public void doCross(SVMClassify svm, String class1Label,
			File[] class1Files, String class2Label, File[] class2Files,
			int kValue, boolean doPredictiveWeights, String outputPath,
			IProgressMonitor monitor, Date dateObj) throws IOException {
		// File folder1 = new File(class1Folder);
		// File folder2 = new File(class2Folder);
		// File[] class1Files = folder1.listFiles();
		// File[] class2Files = folder2.listFiles();
		int numFiles1 = class1Files.length;
		int numFiles2 = class2Files.length;

		int trains1 = (int) Math.floor(0.90 * numFiles1);
		int trains2 = (int) Math.floor(0.90 * numFiles2);

		double[] accuracies = new double[kValue];

		int index1 = 0;
		int index2 = 0;

		for (int i = 1; i <= kValue; i++) {
			ConsoleView.printlInConsoleln("--- Fold " + i + " ---");
			File[] trainFiles1 = new File[trains1];
			File[] trainFiles2 = new File[trains2];
			File[] testFiles1 = new File[numFiles1 - trains1];
			File[] testFiles2 = new File[numFiles2 - trains2];

			int currIndex = index1;
			for (int num = 0; num < trains1; num++) {
				trainFiles1[num] = class1Files[currIndex];
				// ConsoleView.writeInConsole(files1[currIndex]);
				currIndex++;
				if (currIndex >= numFiles1)
					currIndex = 0;
			}
			for (int num = 0; num < numFiles1 - trains1; num++) {
				testFiles1[num] = class1Files[currIndex];
				// ConsoleView.writeInConsole(files1[currIndex]);
				currIndex++;
				if (currIndex >= numFiles1)
					currIndex = 0;
			}

			currIndex = index2;
			for (int num = 0; num < trains2; num++) {
				trainFiles2[num] = class2Files[currIndex];
				// ConsoleView.writeInConsole(files2[currIndex]);
				currIndex++;
				if (currIndex >= numFiles2)
					currIndex = 0;
			}
			for (int num = 0; num < numFiles2 - trains2; num++) {
				testFiles2[num] = class2Files[currIndex];
				// ConsoleView.writeInConsole(files1[currIndex]);
				currIndex++;
				if (currIndex >= numFiles2)
					currIndex = 0;
			}

			svm.cross_train("k" + i, class1Label, trainFiles1, class2Label,
					trainFiles2, doPredictiveWeights,dateObj);
			accuracies[i - 1] = svm.cross_predict("k" + i, class1Label,
					testFiles1, class2Label, testFiles2);

			// Clear required globals like dfmap?
			index1 = index1 + numFiles1 - trains1;
			if (index1 >= numFiles1) {
				index1 = index1 - numFiles1;
			}

			index2 = index2 + numFiles2 - trains2;
			if (index2 >= numFiles2) {
				index2 = index2 - numFiles2;
			}
			monitor.worked(1);
		}

		double averageAccuracy = 0;

		for (int j = 0; j < accuracies.length; j++) {
			averageAccuracy = averageAccuracy + accuracies[j];
		}

		ConsoleView.printlInConsoleln("");
		ConsoleView.printlInConsoleln("Average accuracy over " + kValue
				+ " folds = " + averageAccuracy / accuracies.length + "%");

		clearFiles(kValue, outputPath);
		writeToCSV(accuracies, outputPath,dateObj);
		TacitUtility.createRunReport(outputPath, "SVM Classification",dateObj);
	}

	private void clearFiles(int kValue, String outputPath) {
		ConsoleView.printlInConsoleln("Clearing temporary files");
		for (int i = 0; i < kValue; i++) {
			File toDelete = new File(outputPath
					+ System.getProperty("file.separator")
					+ "SVM_Classification_k" + Integer.toString(i + 1)
					+ ".hashmap");
			toDelete.delete();
			toDelete = new File(outputPath
					+ System.getProperty("file.separator")
					+ "SVM_Classification_k" + Integer.toString(i + 1)
					+ ".model");
			toDelete.delete();
			toDelete = new File(outputPath
					+ System.getProperty("file.separator")
					+ "SVM_Classification_k" + Integer.toString(i + 1) + ".out");
			toDelete.delete();
			toDelete = new File(outputPath
					+ System.getProperty("file.separator")
					+ "SVM_Classification_k" + Integer.toString(i + 1)
					+ ".test");
			toDelete.delete();
			toDelete = new File(outputPath
					+ System.getProperty("file.separator")
					+ "SVM_Classification_k" + Integer.toString(i + 1)
					+ ".train");
			toDelete.delete();
		}
	}

	private void writeToCSV(double[] accuracies, String output, Date dateObj) {
		double averageAccuracy = 0;

		for (int j = 0; j < accuracies.length; j++) {
			averageAccuracy = averageAccuracy + accuracies[j];
		}

		DateFormat df = new SimpleDateFormat("MM-dd-yy-HH-mm-ss");

		String outputPath = output + System.getProperty("file.separator")
				+ "SVM-Classification-" + df.format(dateObj) + ".csv";

		File outFile = new File(outputPath);
		try {
			BufferedWriter bw = new BufferedWriter(new FileWriter(outFile));
			bw.write("Run,Accuracy");
			bw.newLine();

			for (int i = 0; i < accuracies.length; i++) {
				bw.write(Integer.toString(i + 1) + ","
						+ Double.toString(accuracies[i]));
				bw.newLine();
			}
			bw.write("Average accuracy,"
					+ Double.toString(averageAccuracy / accuracies.length));
			bw.close();
			ConsoleView.printlInConsoleln("Finished creating output File - "
					+ outputPath);
		} catch (IOException e) {
			e.printStackTrace();
		}
	}
}