机器学习实践 logistic regression算法

# -- coding: utf-8 --

from numpy import*
from math import log
import operator
import matplotlib.pyplot as plt

def loadDataSet():
    path = r'E:\file\python\test\test\logisticRession_data\testSet.txt'
    dataMat = []
    labelMat = []
    fr = open(path)
    for line in fr.readlines():
        lineArr = line.strip().split()    #移除字符串 头尾 中间的空格
        dataMat.append([1.0, float(lineArr[0]), float(lineArr[1])])  #矩阵增广
        labelMat.append(int(lineArr[2]))
    return dataMat, labelMat

def sigmoid(inX):
    return 1.0/(1 + exp(-inX))

#训练权值
#梯度上升 批处理方法 离线
def gradAscent(dataMatIn, classLabels):
    dataMatrix = mat(dataMatIn)   #构造矩阵 100*3
    labelMat = mat(classLabels).transpose()   #transpose() 转置 变成列向量 100*1
    m, n = shape(dataMatrix)  #m为行数100  n为列数3
    alpha = 0.001     #下降速率参数
    maxCycles = 500   #迭代次数
    weights = ones((n,1))   #构造n，1矩阵  3*1
    for k in range(maxCycles):
        h = sigmoid(dataMatrix * weights)
        error = (labelMat - h)
        weights = weights + alpha*dataMatrix.transpose()*error  #批处理方式
    return weights

#画出决策边界
def plotBestFit(wei):
   # weight = wei.getA()   #将numpy矩阵转换为数组
    weight = wei
    dataMat , labelMat = loadDataSet()
    dataArr = array(dataMat)
    n = shape(dataArr)[0]
    xcord1 = []
    ycord1 = []
    xcord2 = []
    ycord2 = []
    for i in range(n):
        if labelMat[i] == 1:
            xcord1.append(dataArr[i,1])
            ycord1.append(dataArr[i,2])
        else:
            xcord2.append(dataArr[i,1])
            ycord2.append(dataArr[i,2])
    fig = plt.figure()
    ax = fig.add_subplot(111)
    ax.scatter(xcord1, ycord1, s =30, c ='red', marker = 's')
    ax.scatter(xcord2, ycord2, s=30, c='green')
    x = arange(-3.0, 3.0, 0.1)      #与range()类似，返回一个array对象  （起点，终点，步长）
    y = (-weight[0]-weight[1]*x)/weight[2]
    ax.plot(x,y)
    plt.xlabel('X1')
    plt.xlabel('X2')
    plt.show()

#随机梯度上升法  在线算法
def stoGradAscent0(dataMatIn, classLabels):
    m, n = shape(dataMatIn)  # m为行数100  n为列数3
    alpha = 0.01  # 下降速率参数
    weights = ones(n)  # 构造n维向量  数组array
    for i in range(m):
        h = sigmoid(sum(dataMatIn[i] * weights))
        error = classLabels[i] - h
        weights = weights + alpha* error * dataMatIn[i]  #列表list 数组array 矩阵mat 三者不一样  数组可以a*data【i】 列表不可以
    return weights

#随机梯度上升法  改进
def stoGradAscent1(dataMatIn, classLabels, numIter = 150):
    m, n = shape(dataMatIn)  # m为行数100  n为列数3
    weights = ones(n)  # 构造n维向量  数组array
    for j in range(numIter):   #循环150
        dataIndex = range(m)   #范围0-99
        for i in range(m):    #循环100次
            alpha = 4/(1.0+j+i) + 0.01
            randIndex = int(random.uniform(0, len(dataIndex)))
            h = sigmoid(sum(dataMatIn[randIndex] * weights))
            error = classLabels[randIndex] - h
            weights = weights + alpha* error * dataMatIn[randIndex]
            del(dataIndex[randIndex])
    return weights

#分类函数
def classisfyVector(inX, weights):
    prob = sigmoid(sum(inX*weights))
    if prob > 0.5:
        return 1.0
    else:
        return 0.0

path1 = r'E:\file\python\test\test\logisticRession_data\horseColicTest.txt'
path2 = r'E:\file\python\test\test\logisticRession_data\horseColicTraining.txt'

#使用logistic ression 算法
def colicTest():
    frTrain = open(path2)
    frTest = open(path1)
    trainSet = []
    labelSet = []
    for lines in frTrain.readlines():
        curline = lines.strip().split()
        lineArr = []
        for i in range(21):
            lineArr.append(float(curline[i]))
        trainSet.append(lineArr)
        labelSet.append(float(curline[21]))
    wegiht = stoGradAscent1(array(trainSet), labelSet, 500)
    errcount = 0.0
    numtest = 0.0
    for lines in frTest.readlines():
        numtest += 1.0
        curline = lines.strip().split()
        Arr = []
        for i in range(21):
            Arr.append(float(curline[i]))
        if int(classisfyVector(array(Arr), wegiht)) != int(curline[21]):
            errcount += 1.0
    errorrate = errcount/numtest
    print('the error rate is :', errorrate)
    return errorrate

def multitext():
    errorsum = 0.0
    for k in range(10):
        errorsum += colicTest()
    print(errorsum/float(10))

# -- coding: utf-8 --

from numpy import*
from logRegres import*
import matplotlib.pyplot as plt

dataArr, labelMat = loadDataSet()
W = gradAscent(dataArr, labelMat)
print(W)
we0 = stoGradAscent0(array(dataArr), labelMat)  #将dataArr从列表变为数组再带入函数
we1 = stoGradAscent1(array(dataArr), labelMat)
#plotBestFit(we1)


multitext()