当输入为csv格式时,查看Mahout kmeans群集点

时间:2016-02-12 06:23:49

标签: java csv hadoop mahout

您好我正在尝试使用简单的虹膜数据集运行mahout kmeans聚类。我使用以下代码将csv转换为mahout输入

https://github.com/josephmisiti/hadoop-examples/blob/master/mahout/clustering/CSVToMahout.java

代码在将输出提供给kmeans时成功运行。 mahout接受代码并执行提供输出。

现在的问题是让它回到人类可读的格式。

     Configuration config = new Configuration();
        Path path = new Path(args[0]);
        SequenceFile.Reader reader = new SequenceFile.Reader(FileSystem.get(config), path, config);
        DenseVector dv = new DenseVector();

        WritableComparable key = (WritableComparable) reader.getKeyClass().newInstance();

        Writable value = (Writable) reader.getValueClass().newInstance();
        long position = reader.getPosition();
        DataOutput out = null;
        //Cluster cl= new Cluster();

        while(reader.next(key,value))
        {

               System.out.println("Key is: "+key +" value is: "+value+"\n");
                //System.out.println(out.toString());
        }

输出是这样的

键是:0值是:org.apache.mahout.clustering.iterator.ClusterWritable@47b8b32b

键是:1值是:org.apache.mahout.clustering.iterator.ClusterWritable@47b8b32b

键是:2值是:org.apache.mahout.clustering.iterator.ClusterWritable@47b8b32b

提前致谢

0 个答案:

没有答案