Package org.apache.mahout.common.iterator

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

107
108
109
110
111
112
113
114
115
116
117
    MathHelper.assertMatrixEquals(expectedAdjacencyMatrix, actualAdjacencyMatrix);
    MathHelper.assertMatrixEquals(expectedTransitionMatrix, actualTransitionMatrix);

    Map<Integer,Double> rankPerVertex = Maps.newHashMapWithExpectedSize(numVertices);
    for (CharSequence line : new FileLineIterable(new File(outputDir, "part-m-00000"))) {
      String[] tokens = Iterables.toArray(Splitter.on("\t").split(line), String.class);
      rankPerVertex.put(Integer.parseInt(tokens[0]), Double.parseDouble(tokens[1]));
    }

    assertEquals(4, rankPerVertex.size());
View Full Code Here
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
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
    FPGrowthObj<String> fp = new FPGrowthObj<String>();
   
    String inputFilename = "FPGsynth.dat";
    int minSupport = 50;

    StringRecordIterator it = new StringRecordIterator(new FileLineIterable(Resources.getResource(
                                                                                                  inputFilename).openStream()), "\\s+");
    int patternCnt_10_13_1669 = 0;
    int patternCnt_10_13 = 0;
    while (it.hasNext()) {
      Pair<List<String>,Long> next = it.next();
      List<String> items = next.getFirst();
      if (items.contains("10") && items.contains("13")) {
        patternCnt_10_13++;
        if (items.contains("1669")) {
          patternCnt_10_13_1669++;
        }
      }
    }
   
    if (patternCnt_10_13_1669 < minSupport)
      throw new IllegalStateException("the test is broken or data is missing ("
                                      + patternCnt_10_13_1669+", "
                                      + patternCnt_10_13+")");

    final Map<Set<String>,Long> results = Maps.newHashMap();
   
    Set<String> features_10_13 = new HashSet<String>();
    features_10_13.add("10");
    features_10_13.add("13");

    Set<String> returnableFeatures = new HashSet<String>();
    returnableFeatures.add("10");
    returnableFeatures.add("13");
    returnableFeatures.add("1669");
   
    fp.generateTopKFrequentPatterns(
                                    new StringRecordIterator(new FileLineIterable(Resources.getResource(inputFilename).openStream()), "\\s+"),

                                    fp.generateFList(new StringRecordIterator(new FileLineIterable(Resources.getResource(inputFilename)
                                                                                                   .openStream()), "\\s+"), minSupport), minSupport, 100000,
                                    returnableFeatures,
                                    new OutputCollector<String,List<Pair<List<String>,Long>>>() {
       
                                      @Override
View Full Code Here
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
    FPGrowth<String> fp1 = new FPGrowth<String>();

    final 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, 1000000,
                                     returnableFeatures,
                                     new OutputCollector<String,List<Pair<List<String>,Long>>>() {
       
                                       @Override
                                         public void collect(String key, List<Pair<List<String>,Long>> value) {
         
                                         for (Pair<List<String>,Long> v : value) {
                                           List<String> l = v.getFirst();
                                           results1.put(new HashSet<String>(l), v.getSecond());
                                           System.out.println("found pat ["+v.getSecond()+"]: "+ v.getFirst());
                                         }
                                       }
       
                                     }, new StatusUpdater() {
       
                                         @Override
                                           public void update(String status) {}
                                       });

    FPGrowthObj<String> fp2 = new FPGrowthObj<String>();
    final 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, 1000000,
                                     new HashSet<String>(),
                                     new OutputCollector<String,List<Pair<List<String>,Long>>>() {
       
                                       @Override
View Full Code Here
134
135
136
137
138
139
140
141
142
143
144
      for (FileStatus fileStatus : fs.listStatus(verticesPath)) {
        InputStream in = null;
        try {
          in = HadoopUtil.openStream(fileStatus.getPath(), getConf());
          for (String line : new FileLineIterable(in)) {
            writer.append(new IntWritable(index++), new IntWritable(Integer.parseInt(line)));
          }
        } finally {
          Closeables.closeQuietly(in);
        }
View Full Code Here
102
103
104
105
106
107
108
109
110
111
112
    File outputDir2 = getTestTempDir("frequentpatterns2");
    paramsImpl2.set(PFPGrowth.OUTPUT, outputDir2.getAbsolutePath());

    Writer writer = Files.newWriter(input, Charsets.UTF_8);
    try {
      StringRecordIterator it = new StringRecordIterator(new FileLineIterable(Resources.getResource(
        "retail.dat").openStream()), "\\s+");
      Collection<List<String>> transactions = Lists.newArrayList();
     
      while (it.hasNext()) {
        Pair<List<String>,Long> next = it.next();
View Full Code Here
85
86
87
88
89
90
91
92
93
94
95
      itemIDPadded.insert(0, '0');
    }
    List<Preference> prefs = Lists.newArrayList();
    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
75
76
77
78
79
80
81
82
83
84
85
      }

      String messageId = null;
      boolean inBody = false;
      Pattern quotedTextPattern = options.getQuotedTextPattern();
      for (String nextLine : new FileLineIterable(mboxFile, options.getCharset(), false)) {
        if (options.isStripQuotedText() && quotedTextPattern.matcher(nextLine).find()){
          continue;
        }
        for (int i = 0; i < matchers.length; i++) {
          Matcher matcher = matchers[i];
View Full Code Here
237
238
239
240
241
242
243
244
245
246
247
          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
83
84
85
86
87
88
89
90
91
92
  }

  private static List<DummyCandidate> loadPopulation(FileSystem fs, Path f) throws IOException {
    List<DummyCandidate> population = Lists.newArrayList();
    FSDataInputStream in = fs.open(f);
    for (String line : new FileLineIterable(in)) {
      population.add(StringUtils.<DummyCandidate>fromString(line));
    }
    return population;
  }
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.