机器学习实战数据集

机器学习实战数据集

#-*- coding:utf-8 _*-
“““
@author:KING
@file: adaboost.py
@time: 2018/08/15
“““

import numpy as np

def loadSimpData（）:
    datMat = np.matrix（[[1.2.1]
                     [2.1.1]
                     [1.31.]
                     [1.1.]
                     [2.1.]]）
    classLabels = [1.01.0-1.0-1.01.0]
    return datMatclassLabels

def stumpClassify（dataMatrixdimenthreshValthreshIneq）:
    retArray = np.ones（（np.shape（dataMatrix）[0]1））
    if threshIneq == ‘lt‘:
        retArray[dataMatrix[:dimen] <= threshVal] = -1.0
    else:
        retArray[dataMatrix[: dimen] > threshVal] = -1.0
    return retArray

def buildStump（dataArrclassLabelsD）:
    dataMatrix = np.mat（dataArr）
    labelMat = np.mat（classLabels）.T
    mn = np.shape（dataMatrix）
    numSteps = 10.0
    bestStump = {}
    bestClasEst = np.mat（np.zeros（（m1）））
    minError = np.inf
    for i in range（n）:#遍历特征
        rangeMin = dataMatrix[:i].min（）
        rangeMax = dataMatrix[:i].max（）
        stepSize = （rangeMax-rangeMin）/numSteps#阈值增长步长
        for j in range（-1int（numSteps）+1）:
            for inequal in [‘lt‘‘gt‘]:
                threshVal = （rangeMin+float（j）*stepSize）
                predictedVals = stumpClassify（dataMatrixithreshValinequal）
                errArr = np.mat（np.ones（（m1）））
                errArr[predictedVals==labelMat]=0
                weightedError = D.T*errArr
                print（“split:dim %dthresh %.2fthresh inequal:%sthe weightedError is :%.3f“ %（
                    ithreshValinequalweightedError））
                if weightedError                    minError = weightedError
                    bestClasEst = predictedVals.copy（）
                    bestStump[‘dim‘] = i
                    bestStump[‘thresh‘]=threshVal
                    bestStump[‘ineq‘] = inequal
    return bestStumpminErrorbestClasEst

def adaBoostTrainDS（dataArrclassLabelsnumIt = 40）:
    weakClassArr = []
    m = np.shape（dataArr）[0]
    D = np.mat（np.ones（（m1））/m）
    aggClassEst = np.mat（np.zeros（（m1）））
    for i in range（numIt）:
        bestStumperrorclassEst = buildStump（dataArrclassLabelsD）#第i个弱分类器
        print（“D:“D.T）
        alpha = float（0.5*np.log（（1.0-error）/np.fmax（error1e-16）））
        bestStump[‘alpha‘] = alpha
        weakClassArr.append（bestStump）
        print（“classEst: “classEst.T）#该分类器的分类结果
        expon = np.multiply（-1*alpha*np.mat（classLabels）.TclassEst）
        D = np.multiply（Dnp.exp（expon））
        D = D/D.sum（）
        aggClassEst+=alpha*classEst#与此前的分类器同时工作给出分类结果
        print（“aggClassEst: “aggClassEst.T）
        aggErrors = np.multiply（np.sign（aggClassEst）!= np.mat（classLabels）.Tnp.ones（（m1）））
        errorRate = aggErrors.sum（）/m
        print（“total error:“errorRate“\n“）
        if errorRate == 0.0:
            break

    return weakClassArr

def adaClassify（dataToClassclassifierArr）:
    dataMatrix = np.mat（dataToClass）
    m = np.shape（dataMatrix）[0]
    aggClassEst = np.mat（np.

 属性            大小     日期    时间   名称
----------- ---------  ---------- -----  ----
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.idea\
     文件         455  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.idea\AdaBoost.iml
     文件         185  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.idea\misc.xml
     文件         268  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.idea\modules.xml
     文件        6905  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.idea\workspace.xml
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.ipynb_checkpoints\
     文件        6963  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\.ipynb_checkpoints\AdaBoost-checkpoint.ipynb
     文件        8412  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\AdaBoost.ipynb
     文件        3456  2018-08-15 07:18  Machine-Learning-in-Action-master\AdaBoost\adaboost.py
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\.idea\
     文件         398  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\.idea\Logistic回归.iml
     文件         185  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\.idea\misc.xml
     文件         280  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\.idea\modules.xml
     文件        9563  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\.idea\workspace.xml
     文件        3722  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\horseColicTest.txt
     文件       60357  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\horseColicTraining.txt
     文件        4007  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\logRegres.py
     文件        4507  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\logRegres.pyc
     文件        2087  2018-08-15 07:18  Machine-Learning-in-Action-master\Logistic回归\testSet.txt
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\MLiA_SourceCode\
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\MLiA_SourceCode\machinelearninginaction\
     文件          31  2018-08-15 07:18  Machine-Learning-in-Action-master\README.txt
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\SVM\
     目录           0  2018-08-15 07:18  Machine-Learning-in-Action-master\SVM\.idea\
     文件         398  2018-08-15 07:18  Machine-Learning-in-Action-master\SVM\.idea\SVM.iml
     文件         185  2018-08-15 07:18  Machine-Learning-in-Action-master\SVM\.idea\misc.xml
     文件         258  2018-08-15 07:18  Machine-Learning-in-Action-master\SVM\.idea\modules.xml
     文件       16956  2018-08-15 07:18  Machine-Learning-in-Action-master\SVM\.idea\workspace.xml
............此处省略3581个文件信息