这段代码来自《机器学习实战》

注释是我自己加的，英语不好应该有很多语法甚至单词错误，多多包涵
也没什么解释，因为需要注意的都在代码里面了

# k-means algorithm is a algorithm which sort the data that we don't know
# what kind of class it is . we input some data that we know which kind of 
# them , and calculate how far the new data from we knows , then find the 
# nearly class and add the data to this class 
# #
from numpy import *
import operator
'''
    createDataSet Create four data like this

    1.2
    1.1                                                   a
    1.0                                                   a
    0.9
    0.8
    0.7
    0.6
    0.5
    0.4
    0.3
    0.2
    0.1  b
    0.0  b                                                    
        0.0  0.1  0.2  0.3  0.4  0.5  0.6  0.7  0.8  0.9  1.0  1.1  1.2
'''
def createDataSet():
    group = array([[1.0,1.1],[1.0,1.0],[0,0],[0,0.1]])
    labels = ['A','A','B','B']
    return group , labels

group , labels = createDataSet()
print(group)
print(labels)
# #
# about the classify0 function's element
# the first element inX for a vector 
# the secend element dataset for the data which to test in other words 
# the data is 
# 0



def classify0( inX , dataSet , labels , k):
    '''
        1 get the dataSet's row and lines , then , we change the inX to this shape for 
        calculating confortable :: we store the ans to an numpy_array named diffMat
        2 make the diffMat minus the dataSet
        3 square the matrix--diffMat to get Euclidean distance
            about[x , y , z]and[ a , b , c]  we need to calculate the ans as [ (x-a)^2 + (y-b)^2 + (y-c)^2]^0.5
            to get Euclidean distance(欧式距离)
        4 sort the ans
        5 get the most date from ans[0] to ans[k] , create a dic , key = lable value = times
        then get the biggest times's key and print it
    '''
    #  ----start step 1
    dataSetSize = dataSet.shape[0]  
    diffMat = tile(inX , (dataSetSize,1))
    #  ----end step 1

    #  ----start step 2
    diffMat = diffMat - dataSet
    #  -----end step 2

    # --------start step 3
    sqDiffMat = diffMat ** 2
    sqDistances = sqDiffMat.sum(axis = 1)
    distances = sqDistances ** 0.5
    # --------end step 3
    '''
        the ans of distances
        [1.48660687 1.41421356 0.         0.1       ]

        the ans of sortedDistIndicies
        we get the sequence from min to max and store the index to this list 
        named sortedDistIndicies
        [2 3 1 0]
    '''
    # start 4
    sortedDistIndicies = distances.argsort()
    print(sortedDistIndicies)
    # end 4

    # start step 5
    classCount = {}
    print("start for cricyle")
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel , 0) + 1 # the value in dic add
        # print("i"+str(i))
        # print(voteIlabel)
        # print(classCount[voteIlabel])
    print(classCount.items())
    '''
        classCount.items return a list fill of tuple . like this 
        dict_items([('B', 2), ('A', 1)])

        sorted's key means compare with which element ,
        key = operator.itemgetter(1) compare with the secend element

        reverse = True means rise
        reverse = Falst means decline

        at last we get the first value ans[0][0]
    '''
    sortedClassCount = sorted(classCount.items(),
    key = operator.itemgetter(1) , reverse = True)

    return sortedClassCount[0][0]
    # end step 5
print(classify0([0 , 0],group,labels,3))