org.apache.mahout.freqtermsets.FPGrowthDriver.java Source code

Java tutorial

Introduction

Here is the source code for org.apache.mahout.freqtermsets.FPGrowthDriver.java

Source

/**
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements. See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License. You may obtain a copy of the License at
 * 
 * http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.mahout.freqtermsets;

import java.io.File;
import java.io.IOException;
import java.util.Arrays;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

import org.apache.commons.io.FileUtils;
import org.apache.commons.io.FilenameUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.util.ToolRunner;
import org.apache.mahout.common.AbstractJob;
import org.apache.mahout.common.HadoopUtil;
import org.apache.mahout.common.Parameters;
import org.apache.mahout.freqtermsets.PFPGrowth.RunningMode;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import ca.uwaterloo.twitter.TokenIterator;

public final class FPGrowthDriver extends AbstractJob {

    private static final Logger log = LoggerFactory.getLogger(FPGrowthDriver.class);
    private static final String DEFAULT_NUM_THREADS = "1";
    private static final String PARAM_NUM_THREADS = "nJobs";

    private FPGrowthDriver() {
    }

    public static void main(String[] args) throws Exception {
        ToolRunner.run(new Configuration(), new FPGrowthDriver(), args);
    }

    /**
     * Run TopK FPGrowth given the input file,
     */
    @Override
    public int run(String[] args) throws Exception {
        addInputOption();
        addOutputOption();

        addOption("minSupport", "s",
                "(Optional) The minimum number of times a co-occurrence must be present." + " Default Value: 3",
                "3");
        addOption("maxHeapSize", "k",
                "(Optional) Maximum Heap Size k, to denote the requirement to mine top K items."
                        + " Default value: 50",
                "50");
        addOption(PFPGrowth.NUM_GROUPS, "g",
                "(Optional) Number of groups the features should be divided in the map-reduce version."
                        + " Doesn't work in sequential version Default Value:" + PFPGrowth.NUM_GROUPS_DEFAULT,
                Integer.toString(PFPGrowth.NUM_GROUPS_DEFAULT));
        // addOption("splitterPattern", "regex",
        // "Regular Expression pattern used to split given string transaction into"
        // + " itemsets. Default value splits comma separated itemsets.  Default Value:"
        // + " \"[ ,\\t]*[,|\\t][ ,\\t]*\" ", "[ ,\t]*[,|\t][ ,\t]*");
        addOption("numTreeCacheEntries", "tc", "(Optional) Number of entries in the tree cache to prevent duplicate"
                + " tree building. (Warning) a first level conditional FP-Tree might consume a lot of memory, "
                + "so keep this value small, but big enough to prevent duplicate tree building. "
                + "Default Value:5 Recommended Values: [5-10]", "5");
        // addOption("method", "method", "Method of processing: sequential|mapreduce", "mapreduce");
        // //"sequential");
        addOption("encoding", "e", "(Optional) The file encoding.  Default value: UTF-8", "UTF-8");
        // addFlag("useFPG2", "2", "Use an alternate FPG implementation");
        addOption(PFPGrowth.COUNT_IN, "cnt",
                "(Optional) In case of mapreduce, if this is set parallel counting will be skipped and counts will be read from the path specified");
        // addFlag(PFPGrowth.PSEUDO, "ps",
        // "Running on a Pseudo-Cluster (one machine). Uses hardcoded configurations for each job.");
        addOption(PFPGrowth.GROUP_FIS_IN, "gfis",
                "(Optional) In case of mapreduce, if this is set execution will start from the aggregation phase, and group dependent frequent itemsets will be read from the path specified");
        addFlag(AggregatorReducer.MUTUAL_INFO_FLAG, "mi",
                "Set to selec the top K patterns based on the Normalized Mutual Information rather than frequency of pattern");
        addOption(ParallelFPGrowthReducer.MIN_WORDS_FOR_LANG_ID, "lid",
                "The mimun length of a pattern that would be used for language identification");
        addOption(PFPGrowth.MIN_FREQ, "mf",
                "The minimum frequency of a token. Any token with less frequency will be pruned from the begining.");
        addOption(PFPGrowth.PRUNE_PCTILE, "pct",
                "The percentile of frequencies that will be considered; any token with a higher frequency will be pruned");
        //    addFlag("shift", "shift", "If set (and window must be set) it shifts the window by half");
        addFlag(TokenIterator.PARAM_REPEAT_HASHTAG, "rht",
                "If set, each hashtag is repeated, removing the # sign from the second token returned for the same hashtag");
        addOption(PFPGrowth.PARAM_INTERVAL_START, "st",
                "The start time of interval to be mined.. defaults to first known tweet time");
        addOption(PFPGrowth.PARAM_INTERVAL_END, "et",
                "The end time of interval to be mined.. defaults to long.maxvalue");
        addOption(PFPGrowth.PARAM_WINDOW_SIZE, "ws",
                "The duration of windows that will be mined.. defaults to end - start");
        addOption(PFPGrowth.PARAM_STEP_SIZE, "ss",
                "The step by which the window will be advanced.. defaults to windowSize");

        addOption(PARAM_NUM_THREADS, "j",
                "The number of PFP jobs, because in case of intervals resources are under utilized");

        // addOption(PFPGrowth.INDEX_OUT,
        // "ix",
        // "The local folder to which the frequent itemset index will be written");

        if (parseArguments(args) == null) {
            return -1;
        }

        Parameters params = new Parameters();

        if (hasOption("minSupport")) {
            String minSupportString = getOption("minSupport");
            params.set("minSupport", minSupportString);
        }
        if (hasOption("maxHeapSize")) {
            String maxHeapSizeString = getOption("maxHeapSize");
            params.set("maxHeapSize", maxHeapSizeString);
        }
        if (hasOption(PFPGrowth.NUM_GROUPS)) {
            String numGroupsString = getOption(PFPGrowth.NUM_GROUPS);
            params.set(PFPGrowth.NUM_GROUPS, numGroupsString);
        }

        if (hasOption("numTreeCacheEntries")) {
            String numTreeCacheString = getOption("numTreeCacheEntries");
            params.set("treeCacheSize", numTreeCacheString);
        }

        // if (hasOption("splitterPattern")) {
        // String patternString = getOption("splitterPattern");
        // params.set("splitPattern", patternString);
        // }

        String encoding = "UTF-8";
        if (hasOption("encoding")) {
            encoding = getOption("encoding");
        }
        params.set("encoding", encoding);

        // if (hasOption("useFPG2")) {
        // params.set(PFPGrowth.USE_FPG2, "true");
        // }

        // if (hasOption(PFPGrowth.COUNT_IN)) {
        // params.set(PFPGrowth.COUNT_IN, getOption(PFPGrowth.COUNT_IN));
        // }

        // if(hasOption(PFPGrowth.PSEUDO)){
        // params.set(PFPGrowth.PSEUDO, "true");
        // }

        // if (hasOption(PFPGrowth.GROUP_FIS_IN)) {
        // params.set(PFPGrowth.GROUP_FIS_IN, getOption(PFPGrowth.GROUP_FIS_IN));
        // }

        if (hasOption(AggregatorReducer.MUTUAL_INFO_FLAG)) {
            params.set(AggregatorReducer.MUTUAL_INFO_FLAG, "true");
        } else {
            params.set(AggregatorReducer.MUTUAL_INFO_FLAG, "false");
        }

        if (hasOption(ParallelFPGrowthReducer.MIN_WORDS_FOR_LANG_ID)) {
            params.set(ParallelFPGrowthReducer.MIN_WORDS_FOR_LANG_ID,
                    getOption(ParallelFPGrowthReducer.MIN_WORDS_FOR_LANG_ID));
        }

        if (hasOption(PFPGrowth.MIN_FREQ)) {
            params.set(PFPGrowth.MIN_FREQ, getOption(PFPGrowth.MIN_FREQ));
        }

        if (hasOption(PFPGrowth.PRUNE_PCTILE)) {
            params.set(PFPGrowth.PRUNE_PCTILE, getOption(PFPGrowth.PRUNE_PCTILE));
        }

        // if (hasOption(PFPGrowth.PARAM_INTERVAL_END)) {
        params.set(PFPGrowth.PARAM_INTERVAL_END,
                getOption(PFPGrowth.PARAM_INTERVAL_END, Long.toString(Long.MAX_VALUE)));
        // }

        if (hasOption(PFPGrowth.PARAM_WINDOW_SIZE)) {
            params.set(PFPGrowth.PARAM_WINDOW_SIZE, getOption(PFPGrowth.PARAM_WINDOW_SIZE));
        }

        if (hasOption(PFPGrowth.PARAM_STEP_SIZE)) {
            params.set(PFPGrowth.PARAM_STEP_SIZE, getOption(PFPGrowth.PARAM_STEP_SIZE));
        }

        // if (hasOption(PFPGrowth.PARAM_INTERVAL_START)) {
        // params.set(PFPGrowth.PARAM_INTERVAL_START, getOption(PFPGrowth.PARAM_INTERVAL_START));
        // }

        // if (hasOption(PFPGrowth.INDEX_OUT)) {
        // params.set(PFPGrowth.INDEX_OUT, getOption(PFPGrowth.INDEX_OUT));
        // }

        if (hasOption(TokenIterator.PARAM_REPEAT_HASHTAG)) {
            params.set(TokenIterator.PARAM_REPEAT_HASHTAG, "true");
        }

        //    boolean shiftedWindow = hasOption("shift");

        Path inputDir = getInputPath();
        Path outputDir = getOutputPath();

        params.set(PFPGrowth.INPUT, inputDir.toString());
        params.set(PFPGrowth.OUTROOT, outputDir.toString());

        Configuration conf = new Configuration();
        //    HadoopUtil.delete(conf, outputDir);
        FileSystem fs = FileSystem.get(conf);
        if (fs.exists(outputDir)) {
            throw new IllegalArgumentException(
                    "Output path already exists.. please delete it yourself: " + outputDir);
        }

        int nThreads = Integer.parseInt(getOption(PARAM_NUM_THREADS, DEFAULT_NUM_THREADS));
        if (!PFPGrowth.runMode.equals(RunningMode.Batch) && nThreads != 1) {
            throw new UnsupportedOperationException("We use mining results from earlier windows. j must be 1");
        }
        ExecutorService exec = Executors.newFixedThreadPool(nThreads);
        Future<Void> lastFuture = null;

        String startTimeStr = getOption(PFPGrowth.PARAM_INTERVAL_START);
        // params.get(PFPGrowth.PARAM_INTERVAL_START);
        if (startTimeStr == null) {
            // FIXME: Will fail if not running locally.. like many things now
            // FileSystem fs = FileSystem.getLocal(conf);
            // startTimeStr = fs.listStatus(inputDir)[0].getPath().getName();
            File[] startFolders = FileUtils.toFile(inputDir.toUri().toURL()).listFiles();
            Arrays.sort(startFolders);
            startTimeStr = startFolders[0].getName();
        }
        long startTime = Long.parseLong(startTimeStr);
        // Long.toString(PFPGrowth.TREC2011_MIN_TIMESTAMP)));// GMT23JAN2011)));
        long endTime = Long.parseLong(params.get(PFPGrowth.PARAM_INTERVAL_END));
        // Long.toString(Long.MAX_VALUE)));
        long windowSize = Long
                .parseLong(params.get(PFPGrowth.PARAM_WINDOW_SIZE, Long.toString(endTime - startTime)));
        long stepSize = Long.parseLong(params.get(PFPGrowth.PARAM_STEP_SIZE, Long.toString(windowSize)));

        // int numJobs = 0;
        while (startTime < endTime) {
            // if(++numJobs % 100 == 0){
            // Thread.sleep(60000);
            // }
            long shift = 0;
            //      if(shiftedWindow){
            //        shift = (long)Math.floor(windowSize / 2.0f);
            //      }
            params.set(PFPGrowth.PARAM_INTERVAL_START, Long.toString(startTime + shift));

            if (hasOption(PFPGrowth.GROUP_FIS_IN)) {
                String gfisIn = getOption(PFPGrowth.GROUP_FIS_IN);
                gfisIn = FilenameUtils.concat(gfisIn, Long.toString(startTime + shift));
                gfisIn = FilenameUtils.concat(gfisIn,
                        Long.toString(Math.min(endTime, startTime + windowSize) + shift));
                params.set(PFPGrowth.GROUP_FIS_IN, gfisIn);
            }

            if (hasOption(PFPGrowth.COUNT_IN)) {
                String countIn = getOption(PFPGrowth.COUNT_IN);
                //        countIn = FilenameUtils.concat(countIn, Long.toString(startTime + shift));
                //        countIn = FilenameUtils.concat(countIn,
                //            Long.toString(Math.min(endTime, startTime + windowSize) + shift));
                params.set(PFPGrowth.COUNT_IN, countIn);
            }

            String outPathStr = FilenameUtils.concat(outputDir.toString(), Long.toString(startTime + shift));
            outPathStr = FilenameUtils.concat(outPathStr,
                    Long.toString(Math.min(endTime, startTime + windowSize) + shift));
            params.set(PFPGrowth.OUTPUT, outPathStr);

            // PFPGrowth.runPFPGrowth(params);
            lastFuture = exec.submit(new PFPGrowth(params));

            //      startTime += windowSize;
            startTime += stepSize;

            //      Thread.sleep(10000);
        }

        lastFuture.get();
        exec.shutdown();

        while (!exec.isTerminated()) {
            Thread.sleep(1000);
        }

        return 0;
    }

    private static void runFPGrowth(Parameters params) throws IOException {
        throw new UnsupportedOperationException();
        // log.info("Starting Sequential FPGrowth");
        // int maxHeapSize = Integer.valueOf(params.get("maxHeapSize", "50"));
        // int minSupport = Integer.valueOf(params.get("minSupport", "3"));
        //
        // String output = params.get("output", "output.txt");
        //
        // Path path = new Path(output);
        // Configuration conf = new Configuration();
        // FileSystem fs = FileSystem.get(path.toUri(), conf);
        //
        // Charset encoding = Charset.forName(params.get("encoding"));
        // String input = params.get("input");
        //
        // String pattern = params.get("splitPattern", PFPGrowth.SPLITTER.toString());
        //
        // SequenceFile.Writer writer = new SequenceFile.Writer(fs, conf, path, Text.class,
        // TopKStringPatterns.class);
        //
        // if ("true".equals(params.get("useFPG2"))) {
        // org.apache.mahout.freqtermsets.fpgrowth2.FPGrowthObj<String> fp
        // = new org.apache.mahout.freqtermsets.fpgrowth2.FPGrowthObj<String>();
        // Collection<String> features = new HashSet<String>();
        //
        // try {
        // fp.generateTopKFrequentPatterns(
        // new StringRecordIterator(new FileLineIterable(new File(input), encoding, false), pattern),
        // fp.generateFList(
        // new StringRecordIterator(new FileLineIterable(new File(input), encoding, false), pattern),
        // minSupport),
        // minSupport,
        // maxHeapSize,
        // features,
        // new StringOutputConverter(new SequenceFileOutputCollector<Text, TopKStringPatterns>(writer)),
        // new ContextStatusUpdater(null));
        // } finally {
        // Closeables.closeQuietly(writer);
        // }
        // } else {
        // FPGrowth<String> fp = new FPGrowth<String>();
        // Collection<String> features = new HashSet<String>();
        // try {
        // fp.generateTopKFrequentPatterns(
        // new StringRecordIterator(new FileLineIterable(new File(input), encoding, false), pattern),
        // fp.generateFList(
        // new StringRecordIterator(new FileLineIterable(new File(input), encoding, false), pattern),
        // minSupport),
        // minSupport,
        // maxHeapSize,
        // features,
        // new StringOutputConverter(new SequenceFileOutputCollector<Text, TopKStringPatterns>(writer)),
        // new ContextStatusUpdater(null));
        // } finally {
        // Closeables.closeQuietly(writer);
        // }
        // }
        //
        //
        // List<Pair<String, TopKStringPatterns>> frequentPatterns = FPGrowth.readFrequentPattern(conf,
        // path);
        // for (Pair<String, TopKStringPatterns> entry : frequentPatterns) {
        // log.info("Dumping Patterns for Feature: {} \n{}", entry.getFirst(), entry.getSecond());
        // }
    }
}