csc1001 group project part1

#part1 input the data
#data is a list containing all the 2039 training lines as dictionary
import os
dataDir=""
dataFile="D:\\A-Good-Place\\Python\\.vscode\\CSC1001\\src\\train.csv"
def parse_file(datafile):
    data=[]
    with open(datafile,"r",encoding="UTF-8") as f:
        header=f.readline().split(",") #获取表头 header is a list
        counter=0
        for line in f:
            #if counter==1119:
            #   break
            fields=line.split(",")
            entry={}
            for i, value in enumerate(fields):
                entry[header[i].strip()]=value.strip()
            data.append(entry)
            counter+=1
    return data

def splitDataSet(data, axis, value):#去掉进行分类的那个属性
    retDataSet=[]
    for featVec in data:
        if featVec[axis]==value or featVec[axis]>value :
            reducedFeatVec=featVec[:axis]#去掉axis特征
            reducedFeatVec.extend(featVec[axis+1:])
            retDataSet.append(reducedFeatVec)#len（reDataSet）=该feature中的这一类的个数
    return retDataSet



#part2 build the tree
def cart_chooseTheBestFeatureToSplit(data):
    numFeature=len(data[0])-1 #大于或小于6 the last row is not feature
    bestGini=9999
    bestFeature=-1
    for i in range(numFeature):
            featList=[example[i] for example in data]
            uniqueFeature=set(featList) #这个属性中包含的几种元素
            gini=0
            for value in uniqueFeature:
                subDataSet=splitDataSet(data, i, value)#i: 第i个feature； value：i中的某一类
                p=len(subDataSet)/float(len(data))#len（subDataSet）= 该feature中的这一类的个数 len（data）总个数（Dv/D）
                subp=len(splitDataSet(subDataSet,-1,'6'))/float(len(subDataSet))#该feature某类下满足大于6的比例(Gini(Dv))
                gini+=p*(1.0-subp^2-(1-subp)^2)
                if(gini<bestGini):
                    bestGini=gini
                    bestFeature=i
    return bestFeature