bagging算法

from numpy import *
import matplotlib.pyplot as plt
import random
from sklearn import tree
def loadDataSet(fileName):#读取数据
    numFeat = len(open(fileName).readline().split('\t'))
    dataMat = []; labelMat = []
    fr = open(fileName)
    for line in fr.readlines():
        lineArr =[]
        curLine = line.strip().split('\t')
        for i in range(numFeat-1):#添加数据
            lineArr.append(float(curLine[i]))
        dataMat.append(lineArr)
        labelMat.append(float(curLine[-1]))#添加数据对应标签
    return dataMat,labelMat
def rand_train(dataMat,labelMat):#自助法采样
    len_train = len(labelMat)#获取样本1数
    train_data = [] ; train_label = []
    for i in range(len_train):#抽取样本数次样本
        index = random.randint(0,len_train-1)#随机生成样本索引
        train_data.append(dataMat[index])#添加对应数据与标签
        train_label.append(labelMat[index])
    return train_data,train_label#返回训练集与训练集标签
def bagging_by_tree(dataMat,labelMat,t=10):#默认并行生成十个基学习器
    test_data,test_label = loadDataSet('horseColicTest.txt')  #获取测试样本与标签
    predict_list = []
    for i in range(t):#并行生成T个
        train_data,train_label = rand_train(dataMat,labelMat)#自主采样1得到样本
        clf = tree.DecisionTreeClassifier()#初始化决策树模型
        clf.fit(train_data,train_label)#训练模型
        total = []
        y_predicted = clf.predict(test_data)#预测数据
        total.append(y_predicted)
        predict_list.append(total)#结果添加到预测列表中
    return predict_list,test_label
def calc_error(predict_list,test_label):#计算错误率
    m,n,k = shape(predict_list)#提取预测集信息
    predict_label = sum(predict_list,axis = 0)
    predict_label = sign(predict_label)
    for i in range(len(predict_label[0])):
        if predict_label[0][i] == 0:#如果票数相同，则随机生成一个标签
            tip = random.randint(0,1)
            if tip == 0:
                predict_label[0][i] = 1
            else:
                predict_label[0][i] =-1
    error_count = 0#初始化预测错误数
    for i in range(k):
        if predict_label[0][i] != test_label[i]:#判断预测精度
            error_count += 1
    error_rate = error_count/k
    return error_rate
if __name__ == "__main__":
    fileName = 'horseColicTraining.txt'
    dataMat,labelMat =  loadDataSet(fileName)
    train_data,train_label = rand_train(dataMat,labelMat)
    predict_list , test_label = bagging_by_tree(dataMat,labelMat)
    print("Bagging错误率：",calc_error(predict_list,test_label))