jjj.asap.sas.models1.job.BuildBasicMetaCostModels.java Source code

Introduction

Here is the source code for jjj.asap.sas.models1.job.BuildBasicMetaCostModels.java

Source

/*
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU 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 GITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

/*
 * Copyright (C) 2012 James Jesensky
 */

package jjj.asap.sas.models1.job;

import java.io.FileNotFoundException;
import java.io.FileReader;
import java.util.Arrays;
import java.util.HashMap;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import java.util.Queue;
import java.util.concurrent.Future;

import jjj.asap.sas.util.Bucket;
import jjj.asap.sas.util.Contest;
import jjj.asap.sas.util.Job;
import jjj.asap.sas.util.Progress;
import jjj.asap.sas.weka.ModelBuilder;
import weka.attributeSelection.InfoGainAttributeEval;
import weka.attributeSelection.Ranker;
import weka.classifiers.AbstractClassifier;
import weka.classifiers.Classifier;
import weka.classifiers.CostMatrix;
import weka.classifiers.bayes.NaiveBayes;
import weka.classifiers.functions.SMO;
import weka.classifiers.meta.AttributeSelectedClassifier;
import weka.classifiers.meta.Bagging;
import weka.classifiers.meta.CostSensitiveClassifier;
import weka.classifiers.trees.REPTree;

/**
 * Builds a bucket of basic models
 */
public class BuildBasicMetaCostModels extends Job {

    /**
     * args[0] - input bucket
     * args[1] - output bucket
     */
    public static void main(String[] args) throws Exception {
        Job job = new BuildBasicMetaCostModels(args[0], args[1]);
        Job.log("ARGS", Arrays.toString(args));
        job.start();
    }

    private String inputBucket;
    private String outputBucket;

    public BuildBasicMetaCostModels(String inputBucket, String outputBucket) {
        super();
        this.inputBucket = inputBucket;
        this.outputBucket = outputBucket;
    }

    @Override
    protected void run() throws Exception {

        // validate args
        if (!Bucket.isBucket("datasets", inputBucket)) {
            throw new FileNotFoundException(inputBucket);
        }
        if (!Bucket.isBucket("models", outputBucket)) {
            throw new FileNotFoundException(outputBucket);
        }

        // create prototype classifiers
        Map<String, Classifier> prototypes = new HashMap<String, Classifier>();

        // Bagged REPTrees

        Bagging baggedTrees = new Bagging();
        baggedTrees.setNumExecutionSlots(1);
        baggedTrees.setNumIterations(100);
        baggedTrees.setClassifier(new REPTree());
        baggedTrees.setCalcOutOfBag(false);

        prototypes.put("Bagged-REPTrees", baggedTrees);

        // Bagged SMO

        Bagging baggedSVM = new Bagging();
        baggedSVM.setNumExecutionSlots(1);
        baggedSVM.setNumIterations(100);
        baggedSVM.setClassifier(new SMO());
        baggedSVM.setCalcOutOfBag(false);

        prototypes.put("Bagged-SMO", baggedSVM);

        // Meta Cost model for Naive Bayes

        Bagging bagging = new Bagging();
        bagging.setNumExecutionSlots(1);
        bagging.setNumIterations(100);
        bagging.setClassifier(new NaiveBayes());

        CostSensitiveClassifier meta = new CostSensitiveClassifier();
        meta.setClassifier(bagging);
        meta.setMinimizeExpectedCost(true);

        prototypes.put("CostSensitive-MinimizeExpectedCost-NaiveBayes", bagging);

        // init multi-threading
        Job.startService();
        final Queue<Future<Object>> queue = new LinkedList<Future<Object>>();

        // get the input from the bucket
        List<String> names = Bucket.getBucketItems("datasets", this.inputBucket);
        for (String dsn : names) {

            // for each prototype classifier
            for (Map.Entry<String, Classifier> prototype : prototypes.entrySet()) {

                // 
                // speical logic for meta cost
                //

                Classifier alg = AbstractClassifier.makeCopy(prototype.getValue());

                if (alg instanceof CostSensitiveClassifier) {

                    int essaySet = Contest.getEssaySet(dsn);

                    String matrix = Contest.getRubrics(essaySet).size() == 3 ? "cost3.txt" : "cost4.txt";

                    ((CostSensitiveClassifier) alg)
                            .setCostMatrix(new CostMatrix(new FileReader("/asap/sas/trunk/" + matrix)));

                }

                // use InfoGain to discard useless attributes

                AttributeSelectedClassifier classifier = new AttributeSelectedClassifier();

                classifier.setEvaluator(new InfoGainAttributeEval());

                Ranker ranker = new Ranker();
                ranker.setThreshold(0.0001);
                classifier.setSearch(ranker);

                classifier.setClassifier(alg);

                queue.add(Job.submit(
                        new ModelBuilder(dsn, "InfoGain-" + prototype.getKey(), classifier, this.outputBucket)));
            }
        }

        // wait on complete
        Progress progress = new Progress(queue.size(), this.getClass().getSimpleName());
        while (!queue.isEmpty()) {
            try {
                queue.remove().get();
            } catch (Exception e) {
                Job.log("ERROR", e.toString());
            }
            progress.tick();
        }
        progress.done();
        Job.stopService();

    }

}