Run k-means cluster algorithms
Source code name: "clustering_kmeans.py"
Programming language: Python
Topic: Data mining/clustering
DMelt Version 1. Last modified: 12/08/2015. License: Pro
https://datamelt.org/code/cache/clustering_kmeans_1978.py
To run this script using the DataMelt IDE, copy the above URL link to the menu [File]→[Read script from URL] of the DMelt IDE.



from java.util import Random 
from jminhep.algorithms import *
from jminhep.cluster    import *

# create a data holder
data = DataHolder("Example")

# fill 3D data with Gaussian random numbers
rand = Random()
for i in range(100):
      a =[]
      a.append( 10*rand.nextGaussian() )
      a.append( 2*rand.nextGaussian()+1 )
      a.append( 10*rand.nextGaussian()+3 )
      data.add( DataPoint(a) )

# print all outputs      
def printAnswer(alg):   
  print "Name="+alg.getName()
  print "No of final clusters: " + str(alg.getClusters())
  print "No of points in clusters: " + str(alg.getNumberPoints())
  print "Compactness: " + str(alg.getCompactness())
  centers = alg.getCenters() 
  print centers.toString()


# show the data
# HTable(data)
# data.print()

alg=KMeansAlg(data)
alg.setClusters(3)
alg.setOptions(1000,0.001)
alg.run()
printAnswer(alg)

# run 10 times with different seeds
# return the best compactness
alg=KMeansAlg(data)
alg.setClusters(3)
alg.setOptions(1000,0.001)
alg.run(10)
printAnswer(alg)

alg=KMeansExchangeAlg(data)
alg.setClusters(3)
alg.setEpochMax(200)
alg.run()
printAnswer(alg)

Ads help maintain this website.