本文共 4273 字，大约阅读时间需要 14 分钟。

我有联通的2020年扩样后的具体迁徙人数数据，包括所有城市

数据处理代码：

import pandas as pdimport osfrom utils.read_write import eachFile, pdReadCsv'''每个社区到达商圈的平均人口数, #3代表节假日#2代表周末 #1代表工作日    * START_GRID_ID	起始网格编号	string    * START_CITY	起始城市	string    * END_GRID_ID	到达网格编号	string    * END_CITY	到达城市	string    * date	日期	string    * START_TYPE	起始人口类型	string	01-到访 02-居住 03-工作 05职住重合     * END_TYPE	到达人口类型	string	01-到访 02-居住 03-工作    * POP	人数	int    * times 次数'''# def test():#     filepath = os.path.join(root+'000054_0_weekend.txt')#     data = pd.read_csv(filepath, sep='|', usecols=[0, 2, 4, 7], error_bad_lines=False, engine='python')#     column = ['START_GRID_ID', 'END_GRID_ID', 'date', 'pop']#     data.columns = column#     # data = data[data['date'].isin([20191013])]#     workFromCom = pd.merge(data, community, left_on='START_GRID_ID', right_on='YGA_Grid_1', how='right')#     workFromComToMall = pd.merge(workFromCom, mall, left_on='END_GRID_ID', right_on='YGA_Grid_1', how='right')#     workGroup = workFromComToMall.groupby(['SQCODE', 'mall_name']).agg({'pop': sum})#     csv = workGroup['pop'].apply(lambda x: int(x / 5))#     csv.to_csv(filepath + 'holidayFromCommunityToMall.csv', mode='a')def read_file(dirpath):    filepath = os.path.join(dirpath)    print(dirpath)    data = pd.read_csv(filepath, sep='|', usecols=[0, 2, 4, 7], error_bad_lines=False, engine='python')    column = ['START_GRID_ID', 'END_GRID_ID', 'date', 'pop']    data.columns = column    weekend = data[data['date'] == 20191013]    workFromCom = pd.merge(weekend, community, left_on='START_GRID_ID', right_on='YGA_Grid_1', how='right')    workFromComToMall = pd.merge(workFromCom, mall, left_on='END_GRID_ID', right_on='YGA_Grid_1', how='right')    workGroup = workFromComToMall.groupby([ 'SQCODE', 'mall_name']).agg({
   'pop': sum})    csv = workGroup['pop'].apply(lambda x: int(x / 5))    csv.to_csv(save + 'weekendFromCommunityToMall.csv', mode='a',header=False,index=True)    holiday = data[data['date'] < 20191008]    workFromCom = pd.merge(holiday, community, left_on='START_GRID_ID', right_on='YGA_Grid_1', how='right')    workFromComToMall = pd.merge(workFromCom, mall, left_on='END_GRID_ID', right_on='YGA_Grid_1', how='right')    workGroup = workFromComToMall.groupby([ 'SQCODE', 'mall_name']).agg({
   'pop': sum})    csv = workGroup['pop'].apply(lambda x: int(x / 5))    csv.to_csv(save + 'holidayFromCommunityToMall.csv', mode='a',header=False,index=True)    work = data[(data['date'] > 20191007) & (data['date'] != 20191013)]    workFromCom = pd.merge(work, community, left_on='START_GRID_ID', right_on='YGA_Grid_1', how='right')    workFromComToMall = pd.merge(workFromCom, mall, left_on='END_GRID_ID', right_on='YGA_Grid_1', how='right')    workGroup = workFromComToMall.groupby([ 'SQCODE', 'mall_name']).agg({
   'pop': sum})    csv = workGroup['pop'].apply(lambda x: int(x / 5))    csv.to_csv(save + 'workFromCommunityToMall.csv', mode='a',header=False,index=True)def groupby():    src = 'D:\学习文件\项目文件\规土委\data\od\save\save\\'    data = pd.read_csv(src+'workFromCommunityToMall.csv',sep=',',names=['SQCODE','mall_name','pop'])    group = data.groupby(['SQCODE','mall_name']).agg({
   'pop':sum})    csv = group['pop'].apply(lambda x: int(x / 6))    csv.to_csv(src+'workCommunityToMall'+'.csv',header=True)    data = pd.read_csv(src+'holidayFromCommunityToMall.csv',sep=',',names=['SQCODE','mall_name','pop'])    group = data.groupby(['SQCODE','mall_name']).agg({
   'pop':sum})    csv = group['pop'].apply(lambda x: int(x / 7))    csv.to_csv(src+'holidayCommunityToMall'+'.csv',header=True)    data = pd.read_csv(src+'weekendFromCommunityToMall.csv',sep=',',names=['SQCODE','mall_name','pop'])    group = data.groupby(['SQCODE','mall_name']).agg({
   'pop':sum})    group.to_csv(src+'weekendCommunityToMall'+'.csv',header=True)if __name__ == '__main__':    groupby()    root = 'D:\学习文件\项目文件\规土委\data\od\other\\'    save = 'D:\学习文件\项目文件\规土委\data\od\comTomall\\'    grid = 'D:\学习文件\项目文件\规土委\data\od\YGA\\'    community_file = 'com_grid.txt'    community = pdReadCsv(grid + community_file, sep=',')    mall = pd.read_csv(grid + 'mall_grid.txt', sep=',', dtype=str)    # test()    for dir in eachFile(root):        # read_file(root + '000054_0_unholiday')        read_file(root + dir)

转载地址：https://data-mining.blog.csdn.net/article/details/111612516 如侵犯您的版权，请留言回复原文章的地址，我们会给您删除此文章，给您带来不便请您谅解！