Package org.apache.mahout.common.iterator

Examples of org.apache.mahout.common.iterator.FileLineIterable

60
61
62
63
64
65
66
67
68
69
70
    File resultFile = new File(new File(System.getProperty("java.io.tmpdir")), "taste.bookcrossing.txt");
    resultFile.delete();
    PrintWriter writer = null;
    try {
      writer = new PrintWriter(new OutputStreamWriter(new FileOutputStream(resultFile), Charsets.UTF_8));
      for (String line : new FileLineIterable(originalFile, true)) {
        // 0 ratings are basically "no rating", ignore them (thanks h.9000)
        if (line.endsWith("\"0\"")) {
          continue;
        }
        // Delete replace anything that isn't numeric, or a semicolon delimiter. Make comma the delimiter.
View Full Code Here
85
86
87
88
89
90
91
92
93
94
95
      itemIDPadded.insert(0, '0');
    }
    List<Preference> prefs = new ArrayList<Preference>();
    File movieFile = new File(new File(dataDirectory, "training_set"), "mv_00" + itemIDPadded + ".txt");
    try {
      for (String line : new FileLineIterable(movieFile, true)) {
        int firstComma = line.indexOf(',');
        Integer userID = Integer.valueOf(line.substring(0, firstComma));
        int secondComma = line.indexOf(',', firstComma + 1);
        float rating = Float.parseFloat(line.substring(firstComma + 1, secondComma));
        prefs.add(new GenericPreference(userID, itemID, rating));
View Full Code Here
59
60
61
62
63
64
65
66
67
68
69
   
    Map<String, List<String>> byUserEntryCache = new FastMap<String, List<String>>(100000);
   
    for (File byItemFile : byItemDirectory.listFiles()) {
      log.info("Processing {}", byItemFile);
      Iterator<String> lineIterator = new FileLineIterable(byItemFile, false).iterator();
      String line = lineIterator.next();
      String movieIDString = line.substring(0, line.length() - 1);
      while (lineIterator.hasNext()) {
        line = lineIterator.next();
        int firstComma = line.indexOf(',');
View Full Code Here
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
  @Test
  public void testSpecificCaseFromRetailDataMinSup500() throws IOException {
    FPGrowth<String> fp = new FPGrowth<String>();
   
    StringRecordIterator it = new StringRecordIterator(new FileLineIterable(Resources.getResource(
      "retail.dat").openStream()), "\\s+");
    int pattern_41_36_39 = 0;
    while (it.hasNext()) {
      Pair<List<String>,Long> next = it.next();
      List<String> items = next.getFirst();
      if (items.contains("41") && items.contains("36") && items.contains("39")) {
        pattern_41_36_39++;
      }
    }
   
    final Map<Set<String>,Long> results = new HashMap<Set<String>,Long>();
   
    Set<String> returnableFeatures = new HashSet<String>();
    returnableFeatures.add("41");
    returnableFeatures.add("36");
    returnableFeatures.add("39");
   
    fp.generateTopKFrequentPatterns(
      new StringRecordIterator(new FileLineIterable(Resources.getResource("retail.dat").openStream()), "\\s+"),

      fp.generateFList(new StringRecordIterator(new FileLineIterable(Resources.getResource("retail.dat")
          .openStream()), "\\s+"), 500), 500, 1000, returnableFeatures,
      new OutputCollector<String,List<Pair<List<String>,Long>>>() {
       
        @Override
        public void collect(String key, List<Pair<List<String>,Long>> value) {
View Full Code Here
83
84
85
86
87
88
89
90
91
92
  }

  private static List<DummyCandidate> loadPopulation(FileSystem fs, Path f) throws IOException {
    List<DummyCandidate> population = new ArrayList<DummyCandidate>();
    FSDataInputStream in = fs.open(f);
    for (String line : new FileLineIterable(in)) {
      population.add(StringUtils.<DummyCandidate>fromString(line));
    }
    return population;
  }
View Full Code Here
79
80
81
82
83
84
85
86
87
88
89
    assertEquals("files Size: " + files.length + " is not: " + 0, 0, files.length);
    BayesFileFormatter.collapse("animal", analyzer, input, Charsets.UTF_8, new File(out, "animal"));
    files = out.listFiles();
    assertEquals("files Size: " + files.length + " is not: " + 1, 1, files.length);
    int count = 0;
    for (String line : new FileLineIterable(files[0])) {
      assertTrue("line does not start with label", line.startsWith("animal"));
      count++;
    }
    assertEquals(count + " does not equal: " + WORDS.length, count, WORDS.length);
  }
View Full Code Here
233
234
235
236
237
238
239
240
241
242
243
          log.info("Testing: {}", file);
        }
        TimingStatistics operationStats = new TimingStatistics();
       
        long lineNum = 0;
        for (String line : new FileLineIterable(new File(file.getPath()), Charset.forName(params
            .get("encoding")), false)) {
         
          Map<String,List<String>> document = new NGrams(line, Integer.parseInt(params.get("gramSize")))
              .generateNGrams();
          for (Map.Entry<String,List<String>> stringListEntry : document.entrySet()) {
View Full Code Here
58
59
60
61
62
63
64
65
66
67
68
      resultFile.delete();
    }
    Writer writer = null;
    try {
      writer = new OutputStreamWriter(new FileOutputStream(resultFile), Charsets.UTF_8);
      for (String line : new FileLineIterable(originalFile, false)) {
        int lastDelimiterStart = line.lastIndexOf(COLON_DELIMTER);
        if (lastDelimiterStart < 0) {
          throw new IOException("Unexpected input format on line: " + line);
        }
        String subLine = line.substring(0, lastDelimiterStart);
View Full Code Here
60
61
62
63
64
65
66
67
68
69
70
    File resultFile = new File(new File(System.getProperty("java.io.tmpdir")), "taste.bookcrossing.txt");
    resultFile.delete();
    Writer writer = null;
    try {
      writer = new OutputStreamWriter(new FileOutputStream(resultFile), Charsets.UTF_8);
      for (String line : new FileLineIterable(originalFile, true)) {
        // 0 ratings are basically "no rating", ignore them (thanks h.9000)
        if (line.endsWith("\"0\"")) {
          continue;
        }
        // Delete replace anything that isn't numeric, or a semicolon delimiter. Make comma the delimiter.
View Full Code Here
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
      FPGrowth<String> fp1 = new org.apache.mahout.fpm.pfpgrowth.fpgrowth.FPGrowth<String>();

    Map<Set<String>,Long> results1 = Maps.newHashMap();
   
    fp1.generateTopKFrequentPatterns(
      new StringRecordIterator(new FileLineIterable(Resources.getResource(inputFilename).openStream()), "\\s+"),

      fp1.generateFList(new StringRecordIterator(new FileLineIterable(Resources.getResource(inputFilename)
           .openStream()), "\\s+"), minSupport), minSupport, 100000,
      returnableFeatures,
      new MapCollector(results1), new DummyUpdater());

    FPGrowthObj<String> fp2 = new FPGrowthObj<String>();
    Map<Set<String>,Long> initialResults2 = Maps.newHashMap();
    fp2.generateTopKFrequentPatterns(
      new StringRecordIterator(new FileLineIterable(Resources.getResource(inputFilename).openStream()), "\\s+"),

      fp2.generateFList(new StringRecordIterator(new FileLineIterable(Resources.getResource(inputFilename)
           .openStream()), "\\s+"), minSupport), minSupport, 100000,
        Sets.<String>newHashSet(),
      new MapCollector(initialResults2));

    Map<Set<String>, Long> results2;
View Full Code Here
TOP

Related Classes of org.apache.mahout.common.iterator.FileLineIterable

  • mia.recommender.ch05.GenderRescorer
  • org.apache.mahout.cf.taste.example.bookcrossing.BookCrossingDataModel
  • org.apache.mahout.cf.taste.example.grouplens.BooleanDataAndDateDataModel
  • org.apache.mahout.cf.taste.example.grouplens.GroupLensDataModel
  • org.apache.mahout.cf.taste.example.kddcup.track2.TrackItemSimilarity
  • org.apache.mahout.cf.taste.example.netflix.NetflixDataModel
  • org.apache.mahout.cf.taste.example.netflix.NetflixFileDataModel
  • org.apache.mahout.cf.taste.example.netflix.TransposeToByUser
  • org.apache.mahout.cf.taste.hadoop.example.als.netflix.NetflixDatasetConverter
  • org.apache.mahout.cf.taste.hadoop.item.AggregateAndRecommendReducer
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.