字典的增加

最新推荐文章于 2024-01-02 14:55:58 发布

郝锦程

最新推荐文章于 2024-01-02 14:55:58 发布

阅读量361

点赞数

分类专栏：自然语言处理文章标签：词典操作去重 JSON存储 Python 数据处理

本文链接：https://blog.csdn.net/ChangeOline/article/details/125678547

版权

自然语言处理专栏收录该内容

6 篇文章 0 订阅

订阅专栏

if 元素 not in 字典名.keys():
键[元素] = []
键[元素].append(值)

========================================================

import json
‘’’
对总词典去重,然后存储为两个json文件
对总词典去重,
保存词组非单字的
然后存储为两个json文件

‘’’
def len_any(cidui):

front_after = dict()
after_front = dict()
for i in cidui:
    try:
        tmp1,tmp2,tmp3 = i.split(' ')[0].split('\t')
    except:
        continue
    if tmp1 not in front_after.keys():
        front_after[tmp1]=[]
    if tmp2 not in front_after.keys():
        front_after[tmp2]=[]
    front_after[tmp1].append(tmp2)
    front_after[tmp2].append(tmp3)

    if tmp3 not in after_front.keys():
        after_front[tmp3]=[]
    if tmp2 not in after_front.keys():
        after_front[tmp2]=[]
    after_front[tmp3].append(tmp2)
    after_front[tmp2].append(tmp1)
#每个前or后词去重
front_after_res = dict()
for i in front_after.keys():
    front_after_res[i] = tuple(set(front_after[i]))

after_front_res = dict()
for i in after_front.keys():
    after_front_res[i] = tuple(set(after_front[i]))

with open('front_after_res.json', 'w',encoding='utf-8') as f:
    # 设置不转换成ascii  json字符串首缩进
    f.write(json.dumps(front_after_res, ensure_ascii=False))
with open('after_front_res.json', 'w',encoding='utf-8') as f:
    # 设置不转换成ascii  json字符串首缩进
    f.write(json.dumps(after_front_res, ensure_ascii=False))

lentwo_front_after = dict()
for i in front_after.keys():
    if len(i) == 1:
        continue
    tmmp = [j for j in front_after[i] if len(j) > 1]
    if tmmp != []:
        lentwo_front_after[i] = list(set(tmmp))

lentwo_after_front = dict()
for i in after_front.keys():
    if len(i) == 1:
        continue
    tmmp = [j for j in after_front[i] if len(j) > 1]
    if tmmp != []:
        lentwo_after_front[i] = list(set(tmmp))

with open('lentwo_front_after.json', 'w',encoding='utf-8') as f:
    # 设置不转换成ascii
    f.write(json.dumps(lentwo_front_after, ensure_ascii=False,skipkeys=True))
with open('lentwo_after_front.json', 'w',encoding='utf-8') as f:
    # 设置不转换成ascii
    f.write(json.dumps(lentwo_after_front, ensure_ascii=False,skipkeys=True))
print(lentwo_front_after)

cidui=open(‘cidui0505.txt’,‘r’,encoding=‘utf-8’)

len_any(cidui)