Package org.apache.hadoop.conf

Examples of org.apache.hadoop.conf.Configurable

341
342
343
344
345
346
347
348
349
350
351
352
  protected CompressionCodec getInstance(String compressionClass) throws IOException {
    try {
      CompressionCodec codec = (CompressionCodec) Class.forName(compressionClass).newInstance();
      if (codec instanceof Configurable) {
        Configurable configurable = (Configurable) codec;
        configurable.setConf(_configuration);
      }
      return codec;
    } catch (Exception e) {
      throw new IOException(e);
    }
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
                            String measureClassName,
                            double t1,
                            double t2,
                            double convergenceDelta) {
   
    Configurable client = new JobClient();
    JobConf conf = new JobConf(MeanShiftCanopyDriver.class);
   
    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(MeanShiftCanopy.class);
    conf.setJobName("MeanShiftCanopyCluster");
   
    FileInputFormat.setInputPaths(conf, input);
    Path outPath = new Path(output);
    Path controlPath = new Path(input.getParent(), UUID.randomUUID().toString());
    FileOutputFormat.setOutputPath(conf, outPath);
    conf.setMapperClass(MeanShiftCanopyMapper.class);
    conf.setReducerClass(MeanShiftCanopyReducer.class);
    conf.setNumReduceTasks(1);
    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);
    conf.setBoolean("mapred.output.compress", true);
    conf.setClass("mapred.output.compression.codec", GzipCodec.class,  CompressionCodec.class);
    conf.set(MeanShiftCanopyConfigKeys.DISTANCE_MEASURE_KEY, measureClassName);
    conf.set(MeanShiftCanopyConfigKeys.CLUSTER_CONVERGENCE_KEY, String.valueOf(convergenceDelta));
    conf.set(MeanShiftCanopyConfigKeys.T1_KEY, String.valueOf(t1));
    conf.set(MeanShiftCanopyConfigKeys.T2_KEY, String.valueOf(t2));
    conf.set(MeanShiftCanopyConfigKeys.CONTROL_PATH_KEY, controlPath.toString());
    client.setConf(conf);
    try {
      JobClient.runJob(conf);
      input.getFileSystem(conf).delete(input.getParent(), true);
    } catch (IOException e) {
      LOG.warn(e.toString(), e);
View Full Code Here
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
   * @param output
   *          the output pathname String
   */
  public static Path createCanopyFromVectors(String input) {
   
    Configurable client = new JobClient();
    JobConf conf = new JobConf(MeanShiftCanopyDriver.class);
    conf.setJobName("CreateCanopyFromVectors");
   
    conf.setOutputKeyClass(Text.class);
    conf.setOutputValueClass(MeanShiftCanopy.class);
   
    FileInputFormat.setInputPaths(conf, new Path(input));
    Path tmpDir = new Path(new Path(conf.get("hadoop.tmp.dir")), UUID.randomUUID().toString());
    Path outPath = new Path(tmpDir, UUID.randomUUID().toString());
    FileOutputFormat.setOutputPath(conf, outPath);
   
    conf.setMapperClass(MeanShiftCanopyCreatorMapper.class);
    conf.setNumReduceTasks(0);
    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);
   
    client.setConf(conf);
    try {
      JobClient.runJob(conf);
      return outPath;
    } catch (IOException e) {
      LOG.warn(e.toString(), e);
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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
 
  private static final Logger log = LoggerFactory.getLogger(CBayesThetaNormalizerDriver.class);

  @Override
  public void runJob(Path input, Path output, BayesParameters params) throws IOException {
    Configurable client = new JobClient();
    JobConf conf = new JobConf(CBayesThetaNormalizerDriver.class);
    conf.setJobName("Complementary Bayes Theta Normalizer Driver running over input: " + input);
   
    conf.setOutputKeyClass(StringTuple.class);
    conf.setOutputValueClass(DoubleWritable.class);
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-weights/Sigma_j"));
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-tfIdf/trainer-tfIdf"));
    Path outPath = new Path(output, "trainer-thetaNormalizer");
    FileOutputFormat.setOutputPath(conf, outPath);
    // conf.setNumMapTasks(100);
    // conf.setNumReduceTasks(1);
    conf.setMapperClass(CBayesThetaNormalizerMapper.class);
    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setCombinerClass(CBayesThetaNormalizerReducer.class);
    conf.setReducerClass(CBayesThetaNormalizerReducer.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);
    conf.set("io.serializations",
             "org.apache.hadoop.io.serializer.JavaSerialization,"
             + "org.apache.hadoop.io.serializer.WritableSerialization");
    // Dont ever forget this. People should keep track of how hadoop conf
    // parameters and make or break a piece of code
   
    HadoopUtil.delete(conf, outPath);
   
    Path sigmaKFiles = new Path(output, "trainer-weights/Sigma_k/*");
    Map<String,Double> labelWeightSum = SequenceFileModelReader.readLabelSums(sigmaKFiles, conf);
    DefaultStringifier<Map<String,Double>> mapStringifier = new DefaultStringifier<Map<String,Double>>(conf,
        GenericsUtil.getClass(labelWeightSum));
    String labelWeightSumString = mapStringifier.toString(labelWeightSum);
   
    log.info("Sigma_k for Each Label");
    Map<String,Double> c = mapStringifier.fromString(labelWeightSumString);
    log.info("{}", c);
    conf.set("cnaivebayes.sigma_k", labelWeightSumString);
   
    Path sigmaKSigmaJFile = new Path(output, "trainer-weights/Sigma_kSigma_j/*");
    double sigmaJSigmaK = SequenceFileModelReader.readSigmaJSigmaK(sigmaKSigmaJFile, conf);
    DefaultStringifier<Double> stringifier = new DefaultStringifier<Double>(conf, Double.class);
    String sigmaJSigmaKString = stringifier.toString(sigmaJSigmaK);
   
    log.info("Sigma_kSigma_j for each Label and for each Features");
    double retSigmaJSigmaK = stringifier.fromString(sigmaJSigmaKString);
    log.info("{}", retSigmaJSigmaK);
    conf.set("cnaivebayes.sigma_jSigma_k", sigmaJSigmaKString);
   
    Path vocabCountFile = new Path(output, "trainer-tfIdf/trainer-vocabCount/*");
    double vocabCount = SequenceFileModelReader.readVocabCount(vocabCountFile, conf);
    String vocabCountString = stringifier.toString(vocabCount);
   
    log.info("Vocabulary Count");
    conf.set("cnaivebayes.vocabCount", vocabCountString);
    double retvocabCount = stringifier.fromString(vocabCountString);
    log.info("{}", retvocabCount);
    conf.set("bayes.parameters", params.toString());
    conf.set("output.table", output.toString());
    client.setConf(conf);
   
    JobClient.runJob(conf);
   
  }
View Full Code Here
36
37
38
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
*/
public class BayesFeatureDriver implements BayesJob {

  @Override
  public void runJob(Path input, Path output, BayesParameters params) throws IOException {
    Configurable client = new JobClient();
    JobConf conf = new JobConf(BayesFeatureDriver.class);
    conf.setJobName("Bayes Feature Driver running over input: " + input);
    conf.setOutputKeyClass(StringTuple.class);
    conf.setOutputValueClass(DoubleWritable.class);
    conf.setPartitionerClass(FeaturePartitioner.class);
    conf.setOutputKeyComparatorClass(FeatureLabelComparator.class);
    FileInputFormat.setInputPaths(conf, input);
    FileOutputFormat.setOutputPath(conf, output);

    conf.setMapperClass(BayesFeatureMapper.class);

    conf.setInputFormat(KeyValueTextInputFormat.class);
    conf.setCombinerClass(BayesFeatureCombiner.class);
    conf.setReducerClass(BayesFeatureReducer.class);
    conf.setOutputFormat(BayesFeatureOutputFormat.class);
    conf.set("io.serializations",
            "org.apache.hadoop.io.serializer.JavaSerialization,org.apache.hadoop.io.serializer.WritableSerialization");
    // this conf parameter needs to be set enable serialisation of conf values

    HadoopUtil.delete(conf, output);
    conf.set("bayes.parameters", params.toString());

    client.setConf(conf);
    JobClient.runJob(conf);

  }
View Full Code Here
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
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
  private static final Logger log = LoggerFactory.getLogger(BayesTfIdfDriver.class);

  @Override
  public void runJob(Path input, Path output, BayesParameters params) throws IOException {
   
    Configurable client = new JobClient();
    JobConf conf = new JobConf(BayesWeightSummerDriver.class);
    conf.setJobName("TfIdf Driver running over input: " + input);
   
    conf.setOutputKeyClass(StringTuple.class);
    conf.setOutputValueClass(DoubleWritable.class);
   
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-termDocCount"));
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-wordFreq"));
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-featureCount"));
    Path outPath = new Path(output, "trainer-tfIdf");
    FileOutputFormat.setOutputPath(conf, outPath);
   
    // conf.setNumMapTasks(100);
   
    conf.setJarByClass(BayesTfIdfDriver.class);
   
    conf.setMapperClass(BayesTfIdfMapper.class);
    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setCombinerClass(BayesTfIdfReducer.class);
   
    conf.setReducerClass(BayesTfIdfReducer.class);
   
    conf.setOutputFormat(BayesTfIdfOutputFormat.class);
   
    conf.set("io.serializations",
             "org.apache.hadoop.io.serializer.JavaSerialization,"
                 + "org.apache.hadoop.io.serializer.WritableSerialization");
    // Dont ever forget this. People should keep track of how hadoop conf
    // parameters and make or break a piece of code
    HadoopUtil.delete(conf, outPath);
    Path interimFile = new Path(output, "trainer-docCount/part-*");
   
    Map<String,Double> labelDocumentCounts = SequenceFileModelReader.readLabelDocumentCounts(interimFile, conf);
   
    DefaultStringifier<Map<String,Double>> mapStringifier = new DefaultStringifier<Map<String,Double>>(conf,
        GenericsUtil.getClass(labelDocumentCounts));
   
    String labelDocumentCountString = mapStringifier.toString(labelDocumentCounts);
    log.info("Counts of documents in Each Label");
    Map<String,Double> c = mapStringifier.fromString(labelDocumentCountString);
    log.info("{}", c);
   
    conf.set("cnaivebayes.labelDocumentCounts", labelDocumentCountString);
    log.info(params.print());
    conf.set("bayes.parameters", params.toString());
   
    client.setConf(conf);
   
    JobClient.runJob(conf);
  }
View Full Code Here
34
35
36
37
38
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
/** Create and run the Bayes Trainer. */
public class BayesWeightSummerDriver implements BayesJob {

  @Override
  public void runJob(Path input, Path output, BayesParameters params) throws IOException {
    Configurable client = new JobClient();
    JobConf conf = new JobConf(BayesWeightSummerDriver.class);
    conf.setJobName("Bayes Weight Summer Driver running over input: " + input);
   
    conf.setOutputKeyClass(StringTuple.class);
    conf.setOutputValueClass(DoubleWritable.class);
   
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-tfIdf/trainer-tfIdf"));
    Path outPath = new Path(output, "trainer-weights");
    FileOutputFormat.setOutputPath(conf, outPath);
    HadoopUtil.delete(conf, outPath);
    // conf.setNumReduceTasks(1);
    // conf.setNumMapTasks(100);
    conf.setMapperClass(BayesWeightSummerMapper.class);
    // see the javadoc for the spec for file input formats: first token is key,
    // rest is input. Whole document on one line
    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setCombinerClass(BayesWeightSummerReducer.class);
    conf.setReducerClass(BayesWeightSummerReducer.class);
    conf.setOutputFormat(BayesWeightSummerOutputFormat.class);

    conf.set("bayes.parameters", params.toString());
    conf.set("output.table", output.toString());
   
    client.setConf(conf);
   
    JobClient.runJob(conf);
  }
View Full Code Here
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
89
90
91
   *
   * @param params
   *          The Job parameters containing the gramSize, input output folders, defaultCat, encoding
   */
  public static void runJob(Parameters params) throws IOException {
    Configurable client = new JobClient();
    JobConf conf = new JobConf(BayesClassifierDriver.class);
    conf.setJobName("Bayes Classifier Driver running over input: " + params.get("testDirPath"));
    conf.setOutputKeyClass(StringTuple.class);
    conf.setOutputValueClass(DoubleWritable.class);
   
    FileInputFormat.setInputPaths(conf, new Path(params.get("testDirPath")));
    Path outPath = new Path(params.get("testDirPath") + "-output");
    FileOutputFormat.setOutputPath(conf, outPath);
   
    conf.setInputFormat(KeyValueTextInputFormat.class);
    conf.setMapperClass(BayesClassifierMapper.class);
    conf.setCombinerClass(BayesClassifierReducer.class);
    conf.setReducerClass(BayesClassifierReducer.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);
   
    conf.set("io.serializations", "org.apache.hadoop.io.serializer.JavaSerialization,"
                                  + "org.apache.hadoop.io.serializer.WritableSerialization");
   
    HadoopUtil.delete(conf, outPath);
    conf.set("bayes.parameters", params.toString());
   
    client.setConf(conf);
    JobClient.runJob(conf);
   
    Path outputFiles = new Path(outPath, "part*");
    ConfusionMatrix matrix = readResult(outputFiles, conf, params);
    log.info("{}", matrix);
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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
 
  private static final Logger log = LoggerFactory.getLogger(BayesThetaNormalizerDriver.class);

  @Override
  public void runJob(Path input, Path output, BayesParameters params) throws IOException {
    Configurable client = new JobClient();
    JobConf conf = new JobConf(BayesThetaNormalizerDriver.class);
   
    conf.setJobName("Bayes Theta Normalizer Driver running over input: " + input);
   
    conf.setOutputKeyClass(StringTuple.class);
    conf.setOutputValueClass(DoubleWritable.class);
    FileInputFormat.addInputPath(conf, new Path(output, "trainer-tfIdf/trainer-tfIdf"));
    Path outPath = new Path(output, "trainer-thetaNormalizer");
    FileOutputFormat.setOutputPath(conf, outPath);
    // conf.setNumMapTasks(100);
    // conf.setNumReduceTasks(1);
    conf.setMapperClass(BayesThetaNormalizerMapper.class);
    conf.setInputFormat(SequenceFileInputFormat.class);
    conf.setCombinerClass(BayesThetaNormalizerReducer.class);
    conf.setReducerClass(BayesThetaNormalizerReducer.class);
    conf.setOutputFormat(SequenceFileOutputFormat.class);
    conf.set("io.serializations", "org.apache.hadoop.io.serializer.JavaSerialization,"
                                  + "org.apache.hadoop.io.serializer.WritableSerialization");
    // Dont ever forget this. People should keep track of how hadoop conf
    // parameters and make or break a piece of code
   
    HadoopUtil.delete(conf, outPath);

    Path sigmaKFiles = new Path(output, "trainer-weights/Sigma_k/*");
    Map<String,Double> labelWeightSum = SequenceFileModelReader.readLabelSums(sigmaKFiles, conf);
    DefaultStringifier<Map<String,Double>> mapStringifier = new DefaultStringifier<Map<String,Double>>(conf,
        GenericsUtil.getClass(labelWeightSum));
    String labelWeightSumString = mapStringifier.toString(labelWeightSum);
   
    log.info("Sigma_k for Each Label");
    Map<String,Double> c = mapStringifier.fromString(labelWeightSumString);
    log.info("{}", c);
    conf.set("cnaivebayes.sigma_k", labelWeightSumString);
   
    Path sigmaJSigmaKFile = new Path(output, "trainer-weights/Sigma_kSigma_j/*");
    double sigmaJSigmaK = SequenceFileModelReader.readSigmaJSigmaK(sigmaJSigmaKFile, conf);
    DefaultStringifier<Double> stringifier = new DefaultStringifier<Double>(conf, Double.class);
    String sigmaJSigmaKString = stringifier.toString(sigmaJSigmaK);
   
    log.info("Sigma_kSigma_j for each Label and for each Features");
    double retSigmaJSigmaK = stringifier.fromString(sigmaJSigmaKString);
    log.info("{}", retSigmaJSigmaK);
    conf.set("cnaivebayes.sigma_jSigma_k", sigmaJSigmaKString);
   
    Path vocabCountFile = new Path(output, "trainer-tfIdf/trainer-vocabCount/*");
    double vocabCount = SequenceFileModelReader.readVocabCount(vocabCountFile, conf);
    String vocabCountString = stringifier.toString(vocabCount);
   
    log.info("Vocabulary Count");
    conf.set("cnaivebayes.vocabCount", vocabCountString);
    double retvocabCount = stringifier.fromString(vocabCountString);
    log.info("{}", retvocabCount);
    conf.set("bayes.parameters", params.toString());
    conf.set("output.table", output.toString());
    client.setConf(conf);
   
    JobClient.runJob(conf);
   
  }
View Full Code Here
798
799
800
801
802
803
804
805
806
807
    }
  }

  public static <T> T configure(T t) {
    if (t instanceof Configurable) {
      Configurable configurable = (Configurable) t;
      configurable.setConf(new Configuration());
    }
    return t;
  }
View Full Code Here
TOP

Related Classes of org.apache.hadoop.conf.Configurable

  • com.asakusafw.runtime.directio.hadoop.SequenceFileFormat
  • com.fb2pdf.hadoop.cluster.MeanShiftCanopyDriver
  • org.apache.blur.manager.indexserver.DistributedIndexServer
  • org.apache.blur.mapreduce.BlurReducer
  • org.apache.blur.utils.BlurUtil
  • org.apache.hadoop.hive.thrift.HadoopThriftAuthBridge23
  • org.apache.hadoop.thriftfs.HadoopThriftAuthBridge$Server
  • org.apache.mahout.classifier.bayes.mapreduce.bayes.BayesClassifierDriver
  • org.apache.mahout.classifier.bayes.mapreduce.bayes.BayesThetaNormalizerDriver
  • org.apache.mahout.classifier.bayes.mapreduce.cbayes.CBayesThetaNormalizerDriver
    • 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.