Example usage for org.apache.mahout.utils.clustering ClusterWriter write

Introduction

In this page you can find the example usage for org.apache.mahout.utils.clustering ClusterWriter write.

Prototype

void write(ClusterWritable clusterWritable) throws IOException;

Source Link

Document

Write out a Cluster

Usage

From source file:tk.summerway.mahout9.tools.MyClusterDumper.java

License:Apache License

public void printClusters(String[] dictionary) throws Exception {
    Configuration conf = new Configuration();

    if (this.termDictionary != null) {
        if ("text".equals(dictionaryFormat)) {
            dictionary = VectorHelper.loadTermDictionary(new File(this.termDictionary));
        } else if ("sequencefile".equals(dictionaryFormat)) {
            dictionary = VectorHelper.loadTermDictionary(conf, this.termDictionary);
        } else {/*from w ww  .j  a  va 2 s. c  om*/
            throw new IllegalArgumentException("Invalid dictionary format");
        }
    }

    Writer writer;
    boolean shouldClose;
    if (this.outputFile == null) {
        shouldClose = false;
        writer = new OutputStreamWriter(System.out, Charsets.UTF_8);
    } else {
        shouldClose = true;
        if (outputFile.getName().startsWith("s3n://")) {
            Path p = outputPath;
            FileSystem fs = FileSystem.get(p.toUri(), conf);
            writer = new OutputStreamWriter(fs.create(p), Charsets.UTF_8);
        } else {
            Files.createParentDirs(outputFile);
            writer = Files.newWriter(this.outputFile, Charsets.UTF_8);
        }
    }
    ClusterWriter clusterWriter = createClusterWriter(writer, dictionary);
    try {
        long numWritten = clusterWriter.write(new SequenceFileDirValueIterable<ClusterWritable>(
                new Path(seqFileDir, "part-*"), PathType.GLOB, conf));

        writer.flush();
        if (runEvaluation) {
            MyClusterEvaluator ce = new MyClusterEvaluator(pointsDir.toString(), seqFileDir.toString(),
                    "~/cluster_evaluate_result.txt", measure, 1000L);
            ce.evaluateClusters(conf);
        }
        //            if (runEvaluation) {
        //                HadoopUtil.delete(conf, new Path("tmp/representative"));
        //                int numIters = 5;
        //                RepresentativePointsDriver.main(new String[] { "--input",
        //                        seqFileDir.toString(), "--output",
        //                        "tmp/representative", "--clusteredPoints",
        //                        pointsDir.toString(), "--distanceMeasure",
        //                        measure.getClass().getName(), "--maxIter",
        //                        String.valueOf(numIters) });
        //                conf.set(RepresentativePointsDriver.DISTANCE_MEASURE_KEY,
        //                        measure.getClass().getName());
        //                conf.set(RepresentativePointsDriver.STATE_IN_KEY,
        //                        "tmp/representative/representativePoints-" + numIters);
        //                ClusterEvaluator ce = new ClusterEvaluator(conf, seqFileDir);
        //                writer.append("\n");
        //                writer.append("Inter-Cluster Density: ")
        //                        .append(String.valueOf(ce.interClusterDensity()))
        //                        .append("\n");
        //                writer.append("Intra-Cluster Density: ")
        //                        .append(String.valueOf(ce.intraClusterDensity()))
        //                        .append("\n");
        //                CDbwEvaluator cdbw = new CDbwEvaluator(conf, seqFileDir);
        //                writer.append("CDbw Inter-Cluster Density: ")
        //                        .append(String.valueOf(cdbw.interClusterDensity()))
        //                        .append("\n");
        //                writer.append("CDbw Intra-Cluster Density: ")
        //                        .append(String.valueOf(cdbw.intraClusterDensity()))
        //                        .append("\n");
        //                writer.append("CDbw Separation: ")
        //                        .append(String.valueOf(cdbw.separation())).append("\n");
        //                writer.flush();
        //            }
        log.info("Wrote {} clusters", numWritten);
    } finally {
        if (shouldClose) {
            Closeables.close(clusterWriter, false);
        } else {
            if (clusterWriter instanceof GraphMLClusterWriter) {
                clusterWriter.close();
            }
        }
    }
}

---