高校录取分数分析

高校录取分数分析及预测

import pandas as pd
import os
import re

#moveDocLocation 调档分位次 15\16\17的平均值
def avg_moveDocLocation(historydata,year=2018):
    historydata=historydata.fillna(-1)
    historydata["matricDiffYear"]=historydata["matricDiffYear"].astype('int64')
    historydata=historydata.sort_values(by=["collegeCode","order","matricDiffYear"])
    collegecode=historydata["collegeCode"]
    unique_collegecode = collegecode.unique()
    result=pd.DataFrame()
    for code in unique_collegecode:
        temp=historydata.loc[historydata.collegeCode==code,["collegeCode","collegeName","order","matricDiffYear","moveDocLocation","averageLocation","moveDocGrade"]]
        name= temp.loc[0,"collegeName"]
        tt_rst=pd.DataFrame()
        for order in temp["order"].unique():
            temp_order=temp.loc[temp.order==order,:]
            temp_year1 = temp_order.loc[temp_order.matricDiffYear>=(year-3),:]
            temp_year2=temp_year1.loc[temp_year1.matricDiffYear<=(year-1),:]
            temp_year3=temp_year2.loc[temp_year2.moveDocLocation>0,:]
            temp_year4=temp_year3.loc[temp_year3.averageLocation>0,:]
            tmp_value=temp_order.loc[temp_order.matricDiffYear==year,"moveDocGrade"]
            if len(tmp_value)==0:
                tmp_moveDocGrade=0
            else:
                tmp_moveDocGrade=int(tmp_value)
            mean_mov_avg=list(temp_year4[["moveDocLocation","averageLocation"]].mean())
            temp_result = pd.DataFrame([code,name,order,mean_mov_avg[0],mean_mov_avg[1],tmp_moveDocGrade],
                                       index = ["collegecode","collegeName","order","moveDocLocation","averageLocation","moveDocGrade"])
            te