logistic regression编程练习

from numpy import *
#加载模块 numpy
def loadDataSet():
    dataMat = []; labelMat = []
    #加路径的话要写作:open(‘D:\\testSet.txt‘,‘r‘) 缺省为只读
    fr = open(‘testSet.txt‘) 
    #readlines()函数一次读取整个文件,并自动将文本分拆成一个行的列表, 
    #该列表支持python使用for...in...的结构进行处理 (一次只处理一行)   
    for line in fr.readlines():
        #strip()函数 删除字符串中的首尾空格或制表符等  
        #split()函数 按照符号(制表符)进行分割
        lineArr = line.strip().split()
        #每一行加入第零维 x0 = 1
        dataMat.append([1.0, float(lineArr[0]), float(lineArr[1])])
        labelMat.append(int(lineArr[2]))
    return dataMat, labelMat
    
def sigmoid(inX): #定义sigmoid函数
    return 1.0/(1 + exp(-inX))

def gradAscent(dataMatIn, classLabels):
    dataMatrix = mat(dataMatIn)             #转换为numpy内置的矩阵格式
    labelMat = mat(classLabels).transpose() #transpose()是转置的作用
    m,n = shape(dataMatrix)  #获取矩阵的维数
    alpha = 0.001  #设定步长
    maxCycles = 500 #设定循环次数
    weights = ones((n,1)) #初始化权值
    for k in range(maxCycles):              #heavy on matrix operations
        h = sigmoid(dataMatrix*weights)     #logistic regression的hypothesis
        error = (labelMat - h)              
        weights = weights + alpha * dataMatrix.transpose()* error #更新权值
    return weights

>>> import logRegres
>>> dataArr,labelMat = logRegres.loadDataSet()
>>> weights = logRegres.gradAscent(dataArr, labelMat) #原书中漏掉了weights =

def plotBestFit(weights):

    import matplotlib.pyplot as plt #把pyplot重命名为plt, 方便以后使用
    dataMat,labelMat=loadDataSet()
    dataArr = array(dataMat)
    n = shape(dataArr)[0] 
    xcord1 = []; ycord1 = []
    xcord2 = []; ycord2 = []
    for i in range(n):
        if int(labelMat[i])== 1: #标签是1的数据
            xcord1.append(dataArr[i,1]); ycord1.append(dataArr[i,2]) #第一维和第二维分别放入xcorde1和ycorde1这两个list中
        else: #标签是0的数据
            xcord2.append(dataArr[i,1]); ycord2.append(dataArr[i,2]) #第一维和第二维分别放入xcorde2和ycorde2这两个list中
    fig = plt.figure()
    ax = fig.add_subplot(111)
    ax.scatter(xcord1, ycord1, s=30, c=‘red‘, marker=‘s‘) #标签为1的数据标为红色
    ax.scatter(xcord2, ycord2, s=30, c=‘green‘)           #标签为0的数据标为绿色
    x = arange(-3.0, 3.0, 0.1) #其实这里的x = x1, y = x2; 而x0 = 1
    y = (-weights[0]-weights[1]*x)/weights[2] # 0 = weight[0]*x0 + weight[1]*x1 + weight[2]*x2 把分离超平面在二维画出来
    ax.plot(x, y)
    plt.xlabel(‘X1‘); plt.ylabel(‘X2‘);
    plt.show()

def stocGradAscent0(dataMatrix, classLabels):
    m,n = shape(dataMatrix)
    alpha = 0.01 #设定步进值为0.1
    weights = ones(n)   #初始化权值
    for i in range(m): #每次只选取一个点进行权值的更新运算可节省不少时间
        h = sigmoid(sum(dataMatrix[i]*weights))
        error = classLabels[i] - h
        weights = weights + alpha * error * dataMatrix[i]
    return weights

>>> reload(logRegres)

>>> dataArr,labelMat=logRegres.loadDataSet()
>>> weights=logRegres.stocGradAscent0(array(dataArr),labelMat)
>>> logRegres.plotBestFit(weights)

def stocGradAscent1(dataMatrix, classLabels, numIter=150): #可自己设定更新的轮数,默认为150
    m,n = shape(dataMatrix)
    weights = ones(n)   #初始化权值
    for j in range(numIter): #第j轮
        dataIndex = range(m)
        for i in range(m): #第j轮中的第i个数据
            alpha = 4/(1.0+j+i)+0.0001    #alpha会随着更新的次数增加而越来越小
            randIndex = int(random.uniform(0,len(dataIndex)))#每次的i循环的randIndex的值都不同
            h = sigmoid(sum(dataMatrix[randIndex]*weights))
            error = classLabels[randIndex] - h
            weights = weights + alpha * error * dataMatrix[randIndex]
            del(dataIndex[randIndex])
    return weights

def classifyVector(inX, weights): #预测函数
    prob = sigmoid(sum(inX*weights))
    if prob > 0.5: return 1.0
    else: return 0.0

def colicTest():
    frTrain = open(‘horseColicTraining.txt‘); frTest = open(‘horseColicTest.txt‘)
    
    trainingSet = []; trainingLabels = []
    for line in frTrain.readlines(): #训练集有299行
        currLine = line.strip().split(‘\t‘) #每一行的currLine有22个元素
        lineArr =[]
        for i in range(21): #把currLine的前21个元素放入一个list中去
            lineArr.append(float(currLine[i]))
        trainingSet.append(lineArr) # 再把这个list放入一个更大的list中去
        trainingLabels.append(float(currLine[21])) #数据集的最后一列是标签列
    
    trainWeights = stocGradAscent1(array(trainingSet), trainingLabels, 1000) #训练1000轮
    
    errorCount = 0; numTestVec = 0.0
    for line in frTest.readlines(): #测试集有67个数据
        numTestVec += 1.0 #从0开始, 每测试一个,数目加1
        currLine = line.strip().split(‘\t‘)
        lineArr =[] 
        for i in range(21):
            lineArr.append(float(currLine[i])) #生成每个测试数据的list
        if int(classifyVector(array(lineArr), trainWeights))!= int(currLine[21]): #如果预测值与真实值不等
            errorCount += 1 #则错误加1
    errorRate = (float(errorCount)/numTestVec)
    print "the error rate of this test is: %f" % errorRate
    return errorRate

def multiTest():
    numTests = 10; errorSum=0.0
    for k in range(numTests): #测试10次, 求平均
        errorSum += colicTest()
    print "after %d iterations the average error rate is: %f" % (numTests, errorSum/float(numTests))