python统计词频

要求

已知有键值对<店名，城市>的键值对，我们现在的需求是根据城市来统计店的分布。数据的格式如下：

我们希望输出数据的格式如下所示

所有的数据都是以txt文件存储的。

代码如下

# from collections import Counter
# from pprint import pprint

import os
import csv
import codecs

def getNum(l1):
    dic1 = {}
    for i in l1:
        if i in dic1.keys():
            dic1[i] = dic1[i] + 1
        else:
            dic1[i] = 1
    
    return dic1

def main():
    # 统计城市和店名
    f2 = open("pos.txt","r",encoding='utf-8')
    dic = {}
    # dic存储的是[城市，【店1,店2，...】]
    for line in f2:
        # 所有的数据的格式是 [店名（）\t城市\n]
        x = line.split('\t')
        # x的格式应该是 [店名（）\n],[城市]
        y0 = x[0].split('(')
        # 因为数据里面会存在有可能xx，xx(xx路店)，所以我们得只提取店名
        y1 = x[1].split('\n')
        # 提取城市的姓名
        if y1[0] in dic.keys():
            dic[y1[0]].append(y0[0])
        else:
            dic[y1[0]] = []
            dic[y1[0]].append(y0[0])
               
    ans = {}
    dicnew = {}
    with open('params.txt', 'w') as f:
        for key, v in dic.items():
            dic1 = getNum(v)
            f.write(key)
            f.write('\n')
            for k in sorted(dic1,key=dic1.__getitem__):
                f.write(k) 
                f.write(' ')
                f.write(str(dic1[k]))
                f.write('\n')	
  

def main1():
    # 不管城市，统计全国包子店的分布
    f2 = open("pos.txt","r",encoding='utf-8')
    dic = {}
    for line in f2:
        y0 = line.split('(')
        y0 = y0[0]
        if '\n' in y0:
            y0 = y0[0:-1]
        else:
            y0 = y0
        
        if y0 in dic.keys():
            dic[y0] = dic[y0] + 1
        else:   
            dic[y0] =  1
    dicnew = {}        
    for k in sorted(dic,key=dic.__getitem__):
        dicnew[k] = dic[k]
    with open('params.txt', 'w') as f:
        for key, value in dicnew.items():
            f.write(key)
            f.write(' ')
            f.write(str(value))
            f.write('\n')
   

if __name__ == '__main__':
    main()
    # main1()