K近邻算法及其Python实现

 1 from numpy import *
 2 import operator
 3 
 4 def creatDataSet():
 5     group = array([[1.0,1.1],[1.0,1.0],[0,0],[0,0.1]])
 6     labels = [‘A‘,‘A‘,‘B‘,‘B‘]
 7     return group, labels
 8 
 9 def classify(inX,dataSet,labels,k):
10     numSamples = dataSet.shape[0]
11     diffMat = tile(inX,(numSamples,1)) - dataSet
12     sqDiffMat = diffMat**2
13     sqDistances = sqDiffMat.sum(axis = 1)
14     distances = sqDistances ** 0.5
15     sortedDistIndicies = distances.argsort()
16     classCount = {}
17     for i in xrange(k):
18         voteILabel = labels[sortedDistIndicies[i]]
19         classCount[voteILabel] = classCount.get(voteILabel,0) + 1
20     maxLabel = sorted(classCount.iteritems(),
21                       key = operator.itemgetter(1),reverse = True)
22     return maxLabel[0][0]
23 
24 if __name__=="__main__":
25     g,l = creatDataSet()
26     labelOfinX = classify([1.0,1.2],g,l,1)
27     print labelOfinX