数据处理样例

# -*- coding:UTF-8 -*-
import pandas as pd
import numpy as np
import os
os.chdir("D:\Ming\Ma") #数据所在路径

index = ['TIMESTAMP','TA_F','SW_IN_F','VPD_F','P_F','LE_F_MDS', 'LE_CORR','H_F_MDS','H_CORR','GPP_DT_VUT_MEAN','GPP_DT_CUT_MEAN','NEE_VUT_MEAN']  #,'PAR(MJ m-2 d-1)'
sheet = ['20000000','20010000','20020000','20030000','20040000','20050000','20060000','20070000','20080000','20090000','20100000','20110000','20120000','20130000']
for files in os.walk("D:\Ming\Ma"): #数据所在路径
    for file in files[2]:
        data = pd.read_csv(file)
        columns_list = [item for item in index if item in data.columns]
        data_new = data[columns_list]
        data_new['PAR(MJ m-2 d-1)'] = data_new['SW_IN_F']*0.45*0.0864
        data_new = data_new[20000000<=data_new['TIMESTAMP']]
        data_new = data_new[data_new['TIMESTAMP']<=20121231] 
        data_update = data_new.sort_values(['TIMESTAMP'])
        name = file+'result'
        writer = pd.ExcelWriter("D:/Ming/Ma_result/{0}.xlsx".format(name)) #结果所在路径
        data_update.to_excel(writer,'result',index=False)
        #data_update.to_csv("D:/Ming/Ma_result/{0}.csv".format(name),index=False) #结果所在路径
        for i in range(len(sheet)-1):
            data_year = pd.DataFrame()
            data_year = data_update[data_update['TIMESTAMP'] >= int(sheet[i])]
            data_year = data_year[data_update['TIMESTAMP'] < int(sheet[i+1])]
            data_year.to_excel(writer,sheet[i],index=False)

        print 'ok'