Package org.apache.mahout.common

Examples of org.apache.mahout.common.FileLineIterator

308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
   *          data path
   * @return
   */
  public static int[] extractLabels(Dataset dataset, FileSystem fs, Path path) throws IOException {
    FSDataInputStream input = fs.open(path);
    FileLineIterator iterator = new FileLineIterator(input);
   
    int[] labels = new int[dataset.nbInstances()];
    DataConverter converter = new DataConverter(dataset);
   
    int index = 0;
   
    while (iterator.hasNext()) {
      labels[index++] = converter.convert(0, iterator.next()).label;
    }
   
    iterator.close();
   
    return labels;
  }
View Full Code Here
76
77
78
79
80
81
82
83
84
85
86
87
    files = out.listFiles();
    assertEquals("files Size: " + files.length + " is not: " + words.length, files.length, words.length);
    for (File file : files) {
      //should only be one line in the file, and it should be label label
      FileLineIterator it = new FileLineIterator(file);
      String line = it.next().trim();
      assertFalse(it.hasNext());
      String label = "animal" + '\t' + file.getName();
      assertEquals(line + ":::: is not equal to " + label + "::::", line, label);
    }
  }
View Full Code Here
57
58
59
60
61
62
63
64
65
  }
 
  @Override
  protected DataModel buildModel() throws IOException {
    FastByIDMap<Collection<Preference>> data = new FastByIDMap<Collection<Preference>>();
    FileLineIterator iterator = new FileLineIterator(getDataFile(), false);
    processFile(iterator, data, false);
    return new GenericDataModel(GenericDataModel.toDataMap(data, true));
  }
View Full Code Here
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
   * <pre>
   * term DocFreq Index
   * </pre>
   */
  public static String[] loadTermDictionary(InputStream is) throws IOException {
    FileLineIterator it = new FileLineIterator(is);
   
    int numEntries = Integer.parseInt(it.next());
    // System.out.println(numEntries);
    String[] result = new String[numEntries];
   
    while (it.hasNext()) {
      String line = it.next();
      if (line.startsWith("#")) {
        continue;
      }
      String[] tokens = VectorHelper.TAB_PATTERN.split(line);
      if (tokens.length < 3) {
View Full Code Here
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
184
185
186
   
    StringBuilder content = new StringBuilder();
    content.append(header);
    NumberFormat decimalFormatter = new DecimalFormat("0000");
    File dumpFile = new File(dumpFilePath);
    FileLineIterator it;
    if (dumpFilePath.endsWith(".bz2")) {
      // default compression format from http://download.wikimedia.org
      CompressionCodec codec = new BZip2Codec();
      it = new FileLineIterator(codec.createInputStream(new FileInputStream(dumpFile)));
    } else {
      // assume the user has previously de-compressed the dump file
      it = new FileLineIterator(dumpFile);
    }
    int filenumber = 0;
    while (it.hasNext()) {
      String thisLine = it.next();
      if (thisLine.trim().startsWith("<page>")) {
        boolean end = false;
        while (thisLine.trim().startsWith("</page>") == false) {
          content.append(thisLine).append('\n');
          if (it.hasNext()) {
            thisLine = it.next();
          } else {
            end = true;
            break;
          }
        }
View Full Code Here
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
      try {
       
        averageDiffs.clear();
        allRecommendableItemIDs.clear();
       
        FileLineIterator iterator = new FileLineIterator(dataFile, false);
        String firstLine = iterator.peek();
        while ((firstLine.length() == 0) || (firstLine.charAt(0) == COMMENT_CHAR)) {
          iterator.next();
          firstLine = iterator.peek();
        }
        char delimiter = FileDataModel.determineDelimiter(firstLine, 2);
        long averageCount = 0L;
        while (iterator.hasNext()) {
          averageCount = processLine(iterator.next(), delimiter, averageCount);
        }
       
        pruneInconsequentialDiffs();
        updateAllRecommendableItems();
       
View Full Code Here
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
   
    this.dataFile = dataFile.getAbsoluteFile();
    this.lastModified = dataFile.lastModified();
    this.lastUpdateFileModified = readLastUpdateFileModified();
   
    FileLineIterator iterator = new FileLineIterator(dataFile, false);
    String firstLine = iterator.peek();
    while ((firstLine.length() == 0) || (firstLine.charAt(0) == COMMENT_CHAR)) {
      iterator.next();
      firstLine = iterator.peek();
    }
    iterator.close();
    delimiter = determineDelimiter(firstLine, 2);
    hasPrefValues = firstLine.indexOf(delimiter, firstLine.indexOf(delimiter) + 1) >= 0;
   
    this.reloadLock = new ReentrantLock();
    this.transpose = transpose;
View Full Code Here
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
    if (hasPrefValues) {
     
      if (loadFreshData) {
       
        FastByIDMap<Collection<Preference>> data = new FastByIDMap<Collection<Preference>>();
        FileLineIterator iterator = new FileLineIterator(dataFile, false);
        processFile(iterator, data, false);
       
        for (File updateFile : findUpdateFiles()) {
          processFile(new FileLineIterator(updateFile, false), data, false);
        }
       
        return new GenericDataModel(GenericDataModel.toDataMap(data, true));
       
      } else {
       
        FastByIDMap<PreferenceArray> rawData = ((GenericDataModel) delegate).getRawUserData();
       
        for (File updateFile : findUpdateFiles()) {
          processFile(new FileLineIterator(updateFile, false), rawData, true);
        }
       
        return new GenericDataModel(rawData);
       
      }
     
    } else {
     
      if (loadFreshData) {
       
        FastByIDMap<FastIDSet> data = new FastByIDMap<FastIDSet>();
        FileLineIterator iterator = new FileLineIterator(dataFile, false);
        processFileWithoutID(iterator, data);
       
        for (File updateFile : findUpdateFiles()) {
          processFileWithoutID(new FileLineIterator(updateFile, false), data);
        }
       
        return new GenericBooleanPrefDataModel(data);
       
      } else {
       
        FastByIDMap<FastIDSet> rawData = ((GenericBooleanPrefDataModel) delegate).getRawUserData();
       
        for (File updateFile : findUpdateFiles()) {
          processFileWithoutID(new FileLineIterator(updateFile, false), rawData);
        }
       
        return new GenericBooleanPrefDataModel(rawData);
       
      }
View Full Code Here
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
      try {

        averageDiffs.clear();
        allRecommendableItemIDs.clear();
       
        FileLineIterator iterator = new FileLineIterator(dataFile, false);
        String firstLine = iterator.peek();
        while (firstLine.length() == 0 || firstLine.charAt(0) == COMMENT_CHAR) {
          iterator.next();
          firstLine = iterator.peek();
        }
        char delimiter = FileDataModel.determineDelimiter(firstLine);
        long averageCount = 0L;
        while (iterator.hasNext()) {
          averageCount = processLine(iterator.next(), delimiter, averageCount);
        }
       
        pruneInconsequentialDiffs();
        updateAllRecommendableItems();
       
View Full Code Here
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
   * <pre>
   * term DocFreq Index
   * </pre>
   */
  private static String[] loadTermDictionary(InputStream is) throws IOException {
    FileLineIterator it = new FileLineIterator(is);
   
    int numEntries = Integer.parseInt(it.next());
    // System.out.println(numEntries);
    String[] result = new String[numEntries];
   
    while (it.hasNext()) {
      String line = it.next();
      if (line.startsWith("#")) {
        continue;
      }
      String[] tokens = VectorHelper.TAB_PATTERN.split(line);
      if (tokens.length < 3) {
View Full Code Here
TOP

Related Classes of org.apache.mahout.common.FileLineIterator

  • org.apache.mahout.cf.taste.example.jester.JesterDataModel
  • org.apache.mahout.cf.taste.impl.model.file.FileDataModel
  • org.apache.mahout.cf.taste.impl.recommender.slopeone.file.FileDiffStorage
  • org.apache.mahout.cf.taste.impl.similarity.file.FileItemItemSimilarityIterator
  • org.apache.mahout.classifier.bayes.BayesFileFormatterTest
  • org.apache.mahout.classifier.bayes.WikipediaXmlSplitter
  • org.apache.mahout.clustering.lda.LDAPrintTopics
  • org.apache.mahout.df.data.Data
  • org.apache.mahout.utils.clustering.ClusterDumper
  • org.apache.mahout.utils.vectors.VectorHelper
    • 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.