机器学习实战python实例（2）SVM

使用smo算法实现svm
以下代码涉及到的公式推导参照于以下两篇文章，数学推导的部分写的非常好！（如果不了解数学推导过程，代码中的一些部分很可能看不懂）
http://www.thebigdata.cn/JieJueFangAn/12661.html
http://www.wengweitao.com/zhi-chi-xiang-liang-ji-smoxu-lie-zui-xiao-zui-you-hua-suan-fa.html#fnref:calculate
对svm的简要理解可以参见我之前写的http://blog.csdn.net/xiaonannanxn/article/details/52352207

首先我们建立一个SVM.py

# coding:utf-8
from numpy import *
import matplotlib.pyplot as plt

def loadDataSet(filename):
    dataMat = []
    labelMat = []
    fr = open(filename)
    for line in fr.readlines():
        lineArr = line.strip().split('\t')
        dataMat.append([float(lineArr[0]), float(lineArr[1])])
        labelMat.append(float(lineArr[2]))
    return dataMat, labelMat


def selectJrand(i, m):
    j = i
    while j == i:
        j = int(random.uniform(0, m))
    return j


def clipAlpha(aj, H, L):
    if aj > H:
        aj = H
    if aj < L:
        aj = L
    return aj


def smoSimple(dataMatIn, classLabels, C, toler, maxIter):
    dataMatrix = mat(dataMatIn)
    labelMat = mat(classLabels).transpose()
    b = 0
    m, n = dataMatrix.shape
    alphas = mat(zeros((m, 1)))
    Iter = 0
    while Iter < maxIter:
        alphaPairsChanged = 0
        for i in xrange(m):
            # y = wx + b, w = ∑αyx
            fXi = float(multiply(alphas, labelMat).T * dataMatrix * dataMatrix[i, :].T) + b
            Ei = fXi - float(labelMat[i])
            # if α needs to be adjusted or it does not satisfy the ktt
            if ((labelMat[i] * Ei < -toler) and (alphas[i] < C)) or ((labelMat[i] * Ei > toler) and (alphas[i] > 0)):
                j = selectJrand(i, m)
                fXj = float(multiply(alphas, labelMat).T * dataMatrix * dataMatrix[j, :].T) + b
                Ej = fXj - float(labelMat[j])
                alphaIold = alphas[i].copy()
                alphaJold = alphas[j].copy()
                if labelMat[i] != labelMat[j]:
                    L = max(0, alphas[j] - alphas[i])
                    H = min(C, C + alphas[j] - alphas[i])
                else:
                    L = max(0, alphas[j] + alphas[i] - C)
                    H = min(C, alphas[j] + alphas[i])
                if L == H:
                    print "L == H"
                    continue
                eta = 2.0 * dataMatrix[i, :] * dataMatrix[j, :].T \
                      - dataMatrix[i, :] * dataMatrix[i, :].T \
                      - dataMatrix[j, :] * dataMatrix[j, :].T
                if eta >= 0:
                    print "eta >= 0"
                    continue
                alphas[j] -= labelMat[j] * (Ei - Ej) / eta
                alphas[j] = clipAlpha(alphas[j], H, L)
                if abs(alphas[j] - alphaJold) < 0.00001:
                    print "j not moving enough"
                    continue
                alphas[i] += labelMat[j] * labelMat[i] * (alphaJold - alphas[j])
                b1 = b - Ei \
                     - labelMat[i] * (alphas[i] - alphaIold) * dataMatrix[i, :] * dataMatrix[i, :].T \
                     - labelMat[j] * (alphas[j] - alphaJold) * dataMatrix[j, :] * dataMatrix[i, :].T
                b2 = b - Ej \
                     - labelMat[i] * (alphas[i] - alphaIold) * dataMatrix[i, :] * dataMatrix[j, :].T \
                     - labelMat[j] * (alphas[j] - alphaJold) * dataMatrix[j, :] * dataMatrix[j, :].T
                if 0 < alphas[i] < C:
                    b = b1
                elif 0 < alphas[j] < C:
                    b = b2
                else:
                    b = (b1 + b2) / 2.0
                alphaPairsChanged += 1
                print "iter: %d i:%d, pairs changed %d" % (Iter, i, alphaPairsChanged)
        if alphaPairsChanged == 0:
            Iter += 1
        else:
            Iter = 0
        print "iteration number: %d" % Iter
    return b, alphas


def show(dataArr, labelArr, alphas, b):
    for i in xrange(len(labelArr)):
        if labelArr[i] == -1:
            plt.plot(dataArr[i][0], dataArr[i][1], 'or')
        elif labelArr[i] == 1:
            plt.plot(dataArr[i][0], dataArr[i][1], 'Dg')
    # print alphas.shape, mat(labelArr).shape, multiply(alphas, mat(labelArr)).shape
    c = sum(multiply(multiply(alphas.T, mat(labelArr)), mat(dataArr).T), axis=1)
    minY = min(m[1] for m in dataArr)
    maxY = max(m[1] for m in dataArr)
    print minY, maxY
    plt.plot([sum((- b - c[1] * minY) / c[0]), sum((- b - c[1] * maxY) / c[0])], [minY, maxY])
    plt.plot([sum((- b + 1 - c[1] * minY) / c[0]), sum((- b + 1 - c[1] * maxY) / c[0])], [minY, maxY])
    plt.plot([sum((- b - 1 - c[1] * minY) / c[0]), sum((- b - 1 - c[1] * maxY) / c[0])], [minY, maxY])
    plt.show()

以及一个main.py用来测试程序

import SVM
from numpy import *

dataArr, labelArr = SVM.loadDataSet('testSet.txt')
b, alphas = SVM.smoSimple(dataArr, labelArr, 0.6, 0.001, 40)
SVM.show(dataArr, labelArr, alphas, b)

最后可以得到svm的分类结果
我特意增加几个离群的点，以显示出松弛变量对整个分类的影响
蓝线即为分割的超平面，绿线和红线上的点即我们所说的“支持向量”，绿线红线之间的点为离群的点

训练的数据见
http://download.csdn.net/detail/xiaonannanxn/9618859
这次的代码实现了最基本的smo算法，选择αi和αj时分别遍历和随机选择，但是训练100个数据需要14s左右，在增大数据集后这种方法会变得很慢，接下来我会再实现一个优化的smo的算法，并加入核函数