Public class Kmeans {

Public static List FileToVector FileInputStream (fs) throws IOException {
BufferedReader br=new BufferedReader (new InputStreamReader (fs)); 
String readLine=null; 
List Points=new ArrayList (a); 
Double [] data=https://bbs.csdn.net/topics/new double [20]. 
While ((readLine=br. ReadLine ())!=null) {
String [] point=readLine. Split (" "); 
For (int j=0; j<20; J++) 
Data [j]=Double. ParseDouble (point) [j]; 

//double [] fr=data; 
The Vector vec=new RandomAccessSparseVector (data. Length); 
Vec. Assign (data); 
Points. The add (vec); 
} 
Br. The close (); 
Return points; 
} 

Public static void writePointsToFile (List Points, 
The String fileName, 
FileSystem fs, 
The Configuration conf) throws IOException {
Path the Path=new Path (fileName); 

//SequenceFile. Writer (FileSystem, the Configuration, the Path, the key getClass (), the value. The getClass ()) 
SequenceFile. Writer Writer=new SequenceFile. Writer (fs, the conf, 
Path, LongWritable. Class, VectorWritable. Class); 
Long recNum=0; 
VectorWritable vec=new VectorWritable (); 
For (Vector point: points) {
Vec. Set (point); 
Writer. Append (new LongWritable (recNum++), vec); 

} 
Writer. The close (); 
} 
Public static void main (String args []) throws the Exception {

FileInputStream fis=new FileInputStream (" E: \ \ Users \ \ ZZS \ \ Desktop \ \ results \ \ new - data \ \ test. TXT "); 
System. SetProperty (" HADOOP_USER_NAME ", "root"); 
Int k=4; 
List Vectors=FileToVector (fis); 
//List Vectors=getPoints (points); 

The File testData=https://bbs.csdn.net/topics/new File (" clustering/testData "); 
if (! TestData. The exists ()) {
TestData. The mkdir (); 
} 
TestData=https://bbs.csdn.net/topics/new File (" clustering/testData/points "); 
if (! TestData. The exists ()) {
TestData. The mkdir (); 
} 

The Configuration conf=new Configuration (); 
FileSystem fs=FileSystem. Get (conf); 
//has resulted in a store can be mahout recognition vector file1 file 
WritePointsToFile (vectors, "clustering/testdata/points/file1", fs, the conf); 

Path Path=new Path (" clustering/testdata/clusters/part - 00000 "); 

SequenceFile. Writer Writer=new SequenceFile. Writer (fs, the conf, path, the Text. The class, gathering. The class). 

for (int i=0; i //int n=(int) (Math. The random () * points in length). 
The Vector vec=vectors. The get (I); 
Gathering cluster=new gathering (vec, I, new EuclideanDistanceMeasure ());//using Euclidean distance to measure the distance between two points 
Writer. Append (new Text (cluster getIdentifier ()), cluster); 
} 
Writer. The close (); 
KMeansDriver. Run (conf. 
The new Path (" clustering/testdata/points "),//inputPath 
The new Path (" clustering/testdata/clusters "),//ClusterPath 
The new Path (" clustering/output "),//OutputPath 
Convergence coefficient of 0.001,//convergenceDelta 
10,//MaxIterations 
True,//runCLustering 
0,//clusterClassificationThreshold 
True);//runSequential 

SequenceFile. Reader Reader=new SequenceFile. Reader (fs, 
New Path (" clustering/output/" + "clusteredPoints" + "/part - m - 0"), the conf); 


The File f2=new File (" E: \ \ results \ \ hadoop \ \ kmeans "); 
PrintStream ps=new PrintStream (f2 getPath ()); 
//PrintStream Console=System. Out; 
System. SetOut (ps); 


IntWritable key=new IntWritable (); 

WeightedPropertyVectorWritable value=https://bbs.csdn.net/topics/new WeightedPropertyVectorWritable (); 
//ClusterWritable value=https://bbs.csdn.net/topics/new ClusterWritable (); 
While (reader. Next (key, value)) {
System. The out. Println (value. The toString () + "belongs to cluster" + key. The toString ()); 
} 
Ps. The close (); 
Reader. The close (); 
} 

} 

] 

CodePudding user response: