python对TXT文本数据分组统计

txt文档转dataframe写入excel

文本格式：市名-区名-街道名

# -*- coding: utf-8 -*-
import pandas as pd
import numpy as np

df = pd.read_table('no_street.txt')

list=[]
for i in df.values:
    print(i[0])
    li = i[0].split("-")
    # print(li)
    list.append(li)

data = pd.DataFrame(list)
data.to_excel('street.xls')

对excel进行分组统计

street_areas市区重新写入excel

tj_street_areas统计区域和街道个数

# -*- coding: utf-8 -*-
import pandas as pd
import numpy as np

df = pd.read_excel('street.xls')
print(df.head())
print(df.groupby(0).groups)
print(df.loc[74])

for i in df.groupby(0).groups:
    # print(i)
    pass

listm = []
li = []
for name, group in df.groupby(0):
    # print(name) # 市名称
    # print(group)
    column = ['rowid','city','area','street']
    group.columns=column
    # print(group.values)
    cnum = group['area'].nunique()
    snum = group['street'].nunique()
    print(name)
    # print(group['area'])

    for s in group.values:
        print(s)
        li.append(s)
    print("区域：")
    print(group.drop_duplicates(['area'])['area'].values)
    print("街道：")
    print(group.drop_duplicates(['street'])['street'].values)
    # print(cnum)
    # print(snum)
    streets = str(group.drop_duplicates(['street'])['street'].values)
    # s = "市：{}-有{}区-新增街道数：{}个".format(name,cnum,snum)
    areas = str(group.drop_duplicates(['area'])['area'].values)
    s = [name, cnum, snum, areas, streets]
    listm.append(s)
# print(df.groupby(0).get_group('三亚'))#选择某一组数据
# print(df.groupby(0).get_group('三亚')[1].nunique())
# f = open('1.txt', 'w',encoding='utf8')
# f.write(str(listm))
# f.close()

# data = pd.DataFrame(listm)
# data.to_excel('tj_street_areas.xls')
data = pd.DataFrame(li)
data.to_excel('street_areas.xls')