Java 列表<;集群<;DoublePoint>&燃气轮机;设置<;DoublePoint>;
我试图使用DBSCANClusterer(apache.math3)对我生成的一组点进行排序并将其写入文件。在这一点上,我被困在这里:Java 列表<;集群<;DoublePoint>&燃气轮机;设置<;DoublePoint>;,java,data-conversion,dbscan,Java,Data Conversion,Dbscan,我试图使用DBSCANClusterer(apache.math3)对我生成的一组点进行排序并将其写入文件。在这一点上,我被困在这里: public Set<DoublePoint> DBSCAN(Set<DoublePoint> set2) { Set<DoublePoint> points = new Set<DoublePoint>(); DBSCANClusterer<DoublePoint>
public Set<DoublePoint> DBSCAN(Set<DoublePoint> set2) {
Set<DoublePoint> points = new Set<DoublePoint>();
DBSCANClusterer<DoublePoint> dbscan = new DBSCANClusterer<DoublePoint>(1, 15);
//run dbscan on set of points
List<Cluster<DoublePoint>> clusters = dbscan.cluster(set2);
**sorted = clusters???**
publicsetdbscan(setset2){
设置点=新设置();
DBSCANClusterer dbscan=新的DBSCANClusterer(1,15);
//在一组点上运行dbscan
List clusters=dbscan.cluster(set2);
**排序=簇**
如何将:列表簇
分配给设置排序
?我想应该是2D->1D
下面是我剩下的代码:
import java.awt.Point;
import java.io.*;
import java.util.*;
import org.apache.commons.math3.ml.clustering.Cluster;
import org.apache.commons.math3.ml.clustering.DBSCANClusterer;
import org.apache.commons.math3.ml.clustering.DoublePoint;
import java.util.HashSet;
import java.util.Random;
import java.util.Set;
public class Main {
public static void main(final String[] args) throws Exception {
new Main().run();
}
public void run() {
Set<DoublePoint> set = generateSetPoints();
try {
writeToFile(set, "points");
} catch(IOException ex) {
System.out.println("IO Exception while writing to file");
}
Set<DoublePoint> set_by_dbscan = dbScan(set);//
try {
writeToFile(set_by_dbscan, "by_dbscan");
} catch(IOException ex) {
System.out.println("IO Exception while writing to file");
}
}
public Set<DoublePoint> generateSetPoints() {
int xx=100;
int yy=100;
Set<DoublePoint> set = new HashSet<>();
Random rnd = new Random();
int number=100;
do{
int tmp[] = new int[2];
tmp[0] = rnd.nextInt(xx);
tmp[1] = rnd.nextInt(yy);
DoublePoint rndpoint = new DoublePoint(tmp);
set.add(rndpoint);
}
while (set.size()<number);
return set;
}
public void writeToFile(Set<DoublePoint> set, String filename) throws IOException {
File fout = new File( filename + ".txt");
FileOutputStream fos = new FileOutputStream(fout);
BufferedWriter bw = new BufferedWriter(new OutputStreamWriter(fos));
for (DoublePoint p: set) {
bw.write(p.getPoint()[0] + "," + p.getPoint()[1]);
bw.newLine();
}
bw.close();
}
public Set<DoublePoint> dbScan(Set<DoublePoint> set2) {
Set<DoublePoint> points = new Set<DoublePoint>();
DBSCANClusterer dbscan = new DBSCANClusterer(1, 15);
List<Cluster<DoublePoint>> clusters = dbscan.cluster(set2);
return clusters;
}
}
导入java.awt.Point;
导入java.io.*;
导入java.util.*;
导入org.apache.commons.math3.ml.clustering.Cluster;
导入org.apache.commons.math3.ml.clustering.DBSCANClusterer;
导入org.apache.commons.math3.ml.clustering.DoublePoint;
导入java.util.HashSet;
导入java.util.Random;
导入java.util.Set;
公共班机{
公共静态void main(最终字符串[]args)引发异常{
新建Main().run();
}
公开募捐{
Set=generateSetPoints();
试一试{
写入文件(设置“点”);
}捕获(IOEX异常){
System.out.println(“写入文件时的IO异常”);
}
Set Set_by_dbscan=dbscan(Set)//
试一试{
writeToFile(由数据库扫描设置,“由数据库扫描设置”);
}捕获(IOEX异常){
System.out.println(“写入文件时的IO异常”);
}
}
公共集generateSetPoints(){
int xx=100;
int yy=100;
Set=newhashset();
随机rnd=新随机();
整数=100;
做{
int tmp[]=新的int[2];
tmp[0]=rnd.nextInt(xx);
tmp[1]=第二个月(年);
双点rndpoint=新双点(tmp);
set.add(rndpoint);
}
而(set.size()哈希集是未排序的数据结构
如果您希望排序
,请使用保留顺序的东西
sorted=nes HashSet()
是一个真正的WTF
另外,DBSCAN不是用来排序的。请使用光学聚类来代替。谢谢你的回答!因此,我应该使用树集结构来保持顺序。是的,排序后的未排序名称是错误的,但我应该使用DBSCAN来处理我创建、保存和加载的随机生成点。我更新了原始帖子,以包含ude我所有的代码..有什么建议吗?请注意,这是DBSCAN,全大写。我仍然没有得到您想要的“排序”。DBSCAN不排序。通过使用DBSCAN,我想从随机生成的点创建群集。然后将它们保存在输出文件中,并在GNUPLOT中使用它们来可视化它们。您能帮助吗?排序从何而来?我不能为您编写代码。我不使用Apache,ELKI版本要好得多。它有数据生成器和可视化功能,因此您可能不需要编写任何进一步的代码。如果您选中,请不要介意排序。我在提供的代码中更新了我的变量名称。我需要的帮助是将列表转换为集合。