KNN algorithm

KNN.py

#coding=utf-8
#KNN algorithm 
#describe:given a Classification Class={ci|i=1...n} and ci={sample(j)|j=1...n}
#input test_sample 
#output which class the test_sample in/ie.test_sample ∈ C?
import random
def dis(p1,p2):
    f=lambda (x,y):(x-y)**2
    return sum(map(f,zip(p1,p2)))

def classify():
    Class={}
    x=[random.randint(0,100) for i in range(0,40)]
    y=[random.randint(0,100) for i in range(0,40)]
    sample=zip(x,y) #40 samples
    for i in range(4):# 4 class
        Class[i]=sample[10*i:10*(i+1)]
    return Class

if __name__=='__main__':
    #...Initial classification
    Class={}
    Class=classify()
    sample=[]
    for key in Class:
        sample.extend(Class[key])
    test_sample=(50,50)

    #...compute knn  k=10
    k=10
    distance={}
    for s in sample:
        distance[s]=dis(s,test_sample)
    distance=sorted(distance.items(),key=lambda key:key[1])[:k]

    #...vote
    v=[0]*4
    for d in distance:
        if d[0] in Class[0]:
            v[0]+=1
        elif d[0] in Class[1]:
            v[1]+=1
        elif d[0] in Class[2]:
            v[2]+=1
        elif d[0] in Class[3]:
            v[3]+=1
    
    print 'test sample belongs to class %d'%v.index(max(v))