一个k-means算法实例

import numpy as np

def kmeans(X,k,maxIt):
numrow,numlist=X.shape
dataSet=np.zeros((numrow,numlist+1))
dataSet[:, :-1]=X
centroids=dataSet[np.random.randint(numrow,size=k), :]
#centroids=dataSet[0:2, : ]
centroids[:,-1]=range(1,k+1)
iterations=0
oldCentroids=None

while not shouldStop(oldCentroids,centroids,iterations,maxIt):
    oldCentroids=np.copy(centroids)
    iterations+=1
    updateLabels(dataSet,centroids)
    centroids=getCentroids(dataSet,k)

return dataSet

def shouldStop(oldCentroids,centroids,iterations,maxIt):
if iterations>maxIt:
return True
return np.array_equal(oldCentroids,centroids)

def updateLabels(dataSet,centroids):
numrow,numlist=dataSet.shape
for i in range(0,numrow):
dataSet[i,-1]=getLableFromClosestCentroid(dataSet[i, :-1],centroids)

def getLableFromClosestCentroid(dataSetRow,centroid):
label=centroid[0,-1]
minDist=np.linalg.norm(dataSetRow-centroid[0, :-1])
for i in range(1,centroid.shape[0]):
dist=np.linalg.norm(dataSetRow-centroid[i, :-1])
minDist=dist
label=centroid[i,-1]

return label

def getCentroids(dataSet,k):
result=np.zeros((k,dataSet.shape[1]))
for i in range(1,k+1):
oneCluster=dataSet[dataSet[:,-1] == i, :-1]
result[i-1, :-1]=np.mean(oneCluster,axis=0)
result[i-1,-1]=i

return result

x1=np.array([1,1])
x2=np.array([2,1])
x3=np.array([4,3])
x4=np.array([5,4])
testX=np.vstack((x1,x2,x3,x4))
result = kmeans(testX,2,10)

print(result)