KNN 又名K近邻算法
用于：客户流失预测、欺诈检测等。
算法思想：近朱者赤近墨者黑！
步骤：
1. 算距离、计算新数据和训练数据之间的距离 2. 去排序、对算出来的距离进行排序 3. 找邻居、确定最近的K个训练对象 4. 做分类、根据K个近邻对象归属的类最高的为对测试对象分类 # -*- coding: utf-8 -*-"""Created on Thu Aug 3 09:53:27 2017@author: steve"""import syssys.path.append("C:\Users\Administrator.PC-201707110905\Desktop")from numpy import *import operator# 倒入数据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# KNN算法def classify0(Input, training_data, labels, k): training_data_size = training_data.shape[0] InputMatrix = tile(Input, (training_data_size, 1)) - training_data InputMatrix_square = InputMatrix**2 distances = InputMatrix_square.sum(axis=1)**0.5 sorted_distances_indices = distances.argsort() class_count = {} for i in range(k): dict_key = labels[sorted_distances_indices[i]] class_count[dict_key] = class_count.get(dict_key,0) + 1 sorted_class_count = sorted(class_count.items(), key=operator.itemgetter(1), reverse = True) return sorted_class_count[0][0]# 完成这个上面的函数，就可以试试1，2， 分类到哪个啦。# >>> import knn# >>> group, labels = knn.createDataSet()# >>> knn.classify0([1,2], group, labels, 3)# 'A'# 读取数据def file2matrix(filename): fr = open(filename) arrayOnlines = fr.readlines() numberOfLines = len(arrayOnlines) returnMat = zeros((numberOfLines,3)) classLabelVector = [] index = 0 for line in arrayOnlines: line = line.strip() listFromLine = line.split('t') returnMat[index,:] = listFromLine[0:3] classLabelVector.append(int(listFromLine[-1])) index += 1 return returnMat, classLabelVector# 完成这个上面的函数，就可以对把数据转化成需要用的啦# >>> datingDataMat, datingLabels = knn.file2matrix('datingTestSet2.txt')# >>> datingDataMat# array([[ 4.09200000e+04, 8.32697600e+00, 9.53952000e-01], # [ 1.44880000e+04, 7.15346900e+00, 1.67390400e+00], # [ 2.60520000e+04, 1.44187100e+00, 8.05124000e-01], # ..., # [ 2.65750000e+04, 1.06501020e+01, 8.66627000e-01], # [ 4.81110000e+04, 9.13452800e+00, 7.28045000e-01], # [ 4.37570000e+04, 7.88260100e+00, 1.33244600e+00]])# >>> datingLabels[1]# 2# 归一化特征值：因为数值差值较大的属性对计算结果影响较大def autoNorm(dataSet): minVals = dataSet.min(0) maxVals = dataSet.max(0) ranges = maxVals - minVals normDataSet = zeros(shape(dataSet)) m = dataSet.shape[0] normDataSet = dataSet - tile(minVals, 1) normDataSet = normDataSet / tile(ranges, (m, 1)) return normDataSet, ranges, minVals# 完成这个上面的函数，就可以进行归一化特征啦# 特征归一原因是：数字差值较大的属性对结果影响较大# 特征归于的手段：将所有数据转化成0-1的数值# >>> import knn# >>> datingDataMat, datingLabels = knn.file2matrix('datingTestSet2.txt')# >>> normMat,ranges, minVals= knn.autoNorm(datingDataMat)# >>> normMat# array([[ 0.44832535, 0.39805139, 0.56233353], # [ 0.15873259, 0.34195467, 0.98724416], # [ 0.28542943, 0.06892523, 0.47449629], # ..., # [ 0.29115949, 0.50910294, 0.51079493], # [ 0.52711097, 0.43665451, 0.4290048 ], # [ 0.47940793, 0.3768091 , 0.78571804]])# >>> ranges# array([ 9.12730000e+04, 2.09193490e+01, 1.69436100e+00])# >>> minVals# array([ 0. , 0. , 0.001156])# >>> # 分类器针对约会网站得测试代码def datingClassTest(): hoRatio = 0.10 datingDataMat, datingLabels = file2matrix('datingTestSet2.txt') # there is no '2' in intitial book normMat, ranges, minVals = autoNorm(datingDataMat) m = normMat.shape[0] numTestVecs = int(m*hoRatio) errorCount = 0.0 for i in range(numTestVecs): classifierResult = classify0(normMat[i,:], normMat[numTestVecs:m,:],datingLabels[numTestVecs:m], 3) print("the classifier came back with: %d, the real answer is: %d" % (classifierResult, datingLabels[i])) if (classifierResult != datingLabels[i]):errorCount += 1.0 print("the total error rate is: %f" % (errorCount / float(numTestVecs))) print(errorCount) return classifierResult# 完成这个上面的函数，就可以进行测试啦# 测试一般拿90%做训练数据，用10% 做测试数据# >>> import knn# >>> knn.datingClassTest()# the classifier came back with: 3, the real answer is: 3# the classifier came back with: 2, the real answer is: 2# the classifier came back with: 1, the real answer is: 1# the classifier came back with: 1, the real answer is: 1# the classifier came back with: 1, the real answer is: 1# the classifier came back with: 1, the real answer is: 1# .# .# .# the classifier came back with: 1, the real answer is: 1# the classifier came back with: 3, the real answer is: 3# the classifier came back with: 3, the real answer is: 3# the classifier came back with: 2, the real answer is: 2# the classifier came back with: 1, the real answer is: 1# the classifier came back with: 3, the real answer is: 1# the total error rate is: 0.050000# 5.0# 3# >>> # 构建完整可用系统def classifyPerson(): resultList = ['not at all', 'in small doses', 'in large doses'] percentTats = float(input("percentage of time spent playing video games?")) ffMiles = float(input("frequent flier miles earned per years?")) iceCream = float(input("liters of ice cream consumed per year?")) datingDataMat, datingLabels = file2matrix('datingTestSet2.txt') normMat, ranges, minVals = autoNorm(datingDataMat) inArr = array([ffMiles, percentTats, iceCream]) classifierResult = classify0((inArr-minVals)/ranges,normMat,datingLabels, 3) print("you will probably like this person:", resultList[int(classifierResult) - 1])# 完成这个上面的函数 之后就可以进行预测啦！！！# >>> import knn# >>> knn.classifyPerson()# percentage of time spent playing video games?5# frequent flier miles earned per years?5000# liters of ice cream consumed per year?0.2# you will probably like this person: in small doses有关数据的格式咱们来看看
# 对原始数据画出散点图import matplotlibimport matplotlib.pyplot as pltfrom numpy import *fig = plt.figureax = fig.add_subplot(111)datingDataMat, datingLabels = knn.file2matrix('datingTestSet2.txt')ax.scatter(datingDatMat[:,1],datingDatMat[:,2], 15.0*array(datingLabels), 15.0*array[datingLabels])plt.show() 手写字识别系统步骤：
收集数据、提供文本文件准备数据、编写函数img2vector()，将图像格式转换为分类器使用 的向量格式分析数据、python命令行中检查数据，检测可行性测试算法、使用部分数据作为测试样本，给出错误率数据：trainingDigits 2000例子，每个数字有200左右样本。
testDigits 900例子（测试样本）。
稍等啊兄弟姐妹们。。。目前先做另一个项目，那个要的太急。
转载和疑问声明如果你有什么疑问或者想要转载，没有允许是不能转载的哈
赞赏一下能不能转？哈哈，联系我啊，我告诉你呢 ~~
欢迎联系我哈，我会给大家慢慢解答啦~~~怎么联系我？ 笨啊~ ~~ 你留言也行
如果有同学也喜欢的话，欢迎大家加入QQ群交流，群号：533256556；
请备注：csdn
你关注微信公众号1.机器学习算法工程师：2.或者扫那个二维码，后台发送 “我要找朕”，联系我也行啦！
（爱心.gif） 么么哒 ~么么哒 ~么么哒
码字不易啊啊啊，如果你觉得本文有帮助，三毛也是爱！
参考链接