机器学习与python实战（一）-k近邻

kNN（k-nearest neighbor）算法是一个简单而经典的机器学习分类算法，通过度量”待分类数据”和”类别已知的样本”的距离对样本进行分类。

from numpy import *
import operator
#产生数据集
def createDataSet():
    groups = array([[1.0, 1.0], [1.0, 1.1], [0, 0], [0, 0.1]])
    lables = ['A', 'A', 'B', 'B']
    return groups, lables
#分类
def classify0(intX, dataSet, labels, k):
    #1、计算到所有点的距离
    dataSetSize = dataSet.shape[0]
    diffMat = tile(intX, (dataSetSize, 1)) - dataSet
    sqDiffMat = diffMat**2
    sqDistances = sqDiffMat.sum(axis=1)
    distances = sqDistances**0.5
    #2、根据距离进行排序
    sortedClassCount = sqDistances.argsort()
    classCount = {}
    #3、选取种类最多的类别
    for i in range(k):
        voteIlabels = labels[sortedClassCount[i]]
        classCount[voteIlabels] = classCount.get(voteIlabels, 0) + 1
    sortedCount = sorted(classCount.iteritems(), key=operator.itemgetter(1), reverse=True)
    return sortedCount[0][0]
def main():
    groups,labels = createDataSet()
    result = classify0([0, 0], groups, labels, 3)
    print result
main()

来自《机器学习实战》一书，刚刚开始看，目前感觉还比较友好，继续ing