机器学习算法(4) Logistic回归

基于Logistic回归的思想，利用梯度上升的方法，求取回归系数。并且完成对马生病数据的训练和预测。

---

例子来自《Machine Learning in Action》 Peter Harrington

梯度上升

加载数据集

数据集合中有两类共100个数据点
""" 加载数据集 """
def loadDataSet():
    dataMat = []; labelMat = []
    fr = open('testSet.txt')
    for line in fr.readlines():
        lineArr = line.strip().split()
        dataMat.append([1.0, float(lineArr[0]), float(lineArr[1])])
        labelMat.append(int(lineArr[2]))
    return dataMat,labelMat

---

sigmoid函数

利用该函数的函数性质，用于分类

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

---

梯度上升求权重向量

""" 梯度上升 """
def gradAscent(dataMatIn, classLabels):
    dataMatrix = mat(dataMatIn)             #转换为 NumPy 矩阵
    labelMat = mat(classLabels).transpose() #转换为 NumPy 矩阵，求转置 （行向量-->列向量）
    m,n = shape(dataMatrix)                 #获取矩阵的大小
    alpha = 0.001                           #步长
    maxCycles = 500                         #迭代代数
    weights = ones((n,1))                   #权重向量
    for k in range(maxCycles):              
        h = sigmoid(dataMatrix*weights)
        error = (labelMat - h)              # 惩罚度
        weights = weights + alpha * dataMatrix.transpose()* error
    return weights

测试

def testGradAscent():
    dataArr,labelMat = logRegres.loadDataSet()
    weights=logRegres.gradAscent(dataArr,labelMat)
    print(weights)

结果

[[ 4.12414349]
 [ 0.48007329]
 [-0.6168482 ]]

---

可视化结果

""" 绘制拟合后的直线 """
def plotBestFit(weights):
    import matplotlib.pyplot as plt
    dataMat,labelMat=loadDataSet()
    dataArr = array(dataMat)
    n = shape(dataArr)[0]      #  数据点的个数
    xcord1 = []; ycord1 = []
    xcord2 = []; ycord2 = []
    for i in range(n):         #  根据数据点的类型进行分类
        if int(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)
    y = (-weights[0]-weights[1]*x)/weights[2]
    ax.plot(x, y)
    plt.xlabel('X1'); plt.ylabel('X2');
    plt.show()

测试

注意numpy矩阵转换为python数组

"""测试绘制拟合的直线"""
def testPlotBestFit():
    dataArr,labelMat = logRegres.loadDataSet()
    weights =logRegres.gradAscent(dataArr,labelMat)
    logRegres.plotBestFit(weights.getA())    # getA() : matrix --> array

结果

---

随机梯度

之前的梯度计算，当数据集很大的时候，计算量会很大，所以采用随机梯度算法，即每一次迭代只计算一个点。

随机梯度算法_0

""" 随机梯度上升0 """
def stocGradAscent0(dataMatrix, classLabels):
    m,n = shape(dataMatrix)
    alpha = 0.01
    weights = ones(n)   
    for i in range(m):
        h = sigmoid(sum(dataMatrix[i]*weights)) # 每次只选取一个特征点进行训练
        error = classLabels[i] - h
        weights = weights + alpha * error * dataMatrix[i]
    return weights

测试

"""测试随机梯度上升0"""
def teststocGradAscent0():
    dataArr,labelMat = logRegres.loadDataSet()
    weights =logRegres.stocGradAscent0(array(dataArr),labelMat)
    logRegres.plotBestFit(weights)

结果

由于迭代次数比较少，所以划分效果不是很理想。

---

随机梯度算法_1

分析之前效果不理想的原因：
1. 由于迭代过程中步长固定，所以在最后收敛的过程中，会周期震荡。
2. 每次的训练点不是随机取得，会收到数据周期性的影响。

针对这两点，做出如下修改：
1. alpha 步长大小随着迭代的次数而减少
2. 随机选取数据点进行训练

""" 随机梯度上升1 """
def stocGradAscent1(dataMatrix, classLabels, numIter=150):
    m,n = shape(dataMatrix)
    weights = ones(n)   
    for j in range(numIter):
        dataIndex = list(range(m))    # rang 对象无法迭代
        for i in range(m):
            alpha = 4/(1.0+j+i)+0.0001    # 步长会随着迭代进行而减少，但不会为0。防止波动和停止不前
            randIndex = int(random.uniform(0,len(dataIndex)))  # 随机选取迭代值，防止周期波动
            h = sigmoid(sum(dataMatrix[randIndex]*weights))
            error = classLabels[randIndex] - h
            weights = weights + alpha * error * dataMatrix[randIndex]
            del(dataIndex[randIndex])
    return weights

测试

"""测试随机梯度上升1"""
def teststocGradAscent1():
    dataArr,labelMat = logRegres.loadDataSet()
    weights =logRegres.stocGradAscent1(array(dataArr),labelMat)
    logRegres.plotBestFit(weights)

结果

可以看到，这次的划分效果就很好了。

应用

利用Logistic回归来预测病马的死亡率

训练

""" 利用回归系数和特征量计算类别"""
def classifyVector(inX, weights):
    prob = sigmoid(sum(inX*weights))
    if prob > 0.5: return 1.0
    else: return 0.0

""" 加载数据 训练 测试"""
def colicTest():
    # 训练回归系数
    frTrain = open('horseColicTraining.txt'); frTest = open('horseColicTest.txt')
    trainingSet = []; trainingLabels = []
    for line in frTrain.readlines():
        currLine = line.strip().split('\t')
        lineArr =[]
        for i in range(21):
            lineArr.append(float(currLine[i]))
        trainingSet.append(lineArr)
        trainingLabels.append(float(currLine[21]))
    trainWeights = stocGradAscent1(array(trainingSet), trainingLabels, 500)
    # 测试分类效果
    errorCount = 0; numTestVec = 0.0
    for line in frTest.readlines():
        numTestVec += 1.0
        currLine = line.strip().split('\t')
        lineArr =[]
        for i in range(21):
            lineArr.append(float(currLine[i]))
        if int(classifyVector(array(lineArr), trainWeights))!= int(currLine[21]):
            errorCount += 1
    errorRate = (float(errorCount)/numTestVec)
    print ("the error rate of this test is: %f" % errorRate)
    return errorRate

测试

"""预测病马死亡率"""    
def multiTest():
    numTests = 10; errorSum=0.0
    for k in range(numTests):
        errorSum += logRegres.colicTest()
    print ("after %d iterations the average error rate is: %f" % (numTests, errorSum/float(numTests)))

结果

the error rate of this test is: 0.432836
the error rate of this test is: 0.268657
the error rate of this test is: 0.417910
the error rate of this test is: 0.313433
the error rate of this test is: 0.298507
the error rate of this test is: 0.358209
the error rate of this test is: 0.298507
the error rate of this test is: 0.283582
the error rate of this test is: 0.388060
the error rate of this test is: 0.402985
after 10 iterations the average error rate is: 0.346269

---

以上完整代码见GitHub。