python构建 城市和省份字典 的实例应用

UserInfoData1 = pd.read_csv(‘data.csv‘,encoding = ‘GBK‘)
UserInfoData1.head()

# 去掉城市名称中的 “市”
# UserInfoData1.UserInfo_20 = [x[0:-1] for x in UserInfoData1.UserInfo_20 if UserInfoData1.UserInfo_20.notnull().all()]
# Length of values does not match length of index

# 用此方法的时候，当把序列里面的单个缺失值拿出来的时候，不能判断是否为缺失值，因此转化为字符串判断 == ‘nan’
def delete_shi(s):
    str_s = str(s)
    if str_s==‘nan‘:          # AttributeError: ‘str‘ object has no attribute ‘isnull‘
        return s
    else:
        if str_s[-1]==‘市‘:
            return str_s[0:-1]
        else:
            return s
    
UserInfoData1.UserInfo_20 = UserInfoData1.UserInfo_20.apply(lambda s: delete_shi(s))
UserInfoData1.UserInfo_8 = UserInfoData1.UserInfo_20.apply(lambda s: delete_shi(s))

# UserInfo_2 为非缺失值
index=UserInfoData1[UserInfoData1.UserInfo_2.notnull() & UserInfoData1.UserInfo_4.isnull() & UserInfoData1.UserInfo_8.isnull() & UserInfoData1.UserInfo_20.isnull()].index
UserInfoData1[UserInfoData1.UserInfo_2.notnull() & UserInfoData1.UserInfo_4.isnull() & UserInfoData1.UserInfo_8.isnull() & UserInfoData1.UserInfo_20.isnull()]

# 填补UserInfo_4缺失值
UserInfoData1.UserInfo_4[index]=UserInfoData1.UserInfo_2[index]

# 处理省份变量
def delete_postfix1(s,str):
    if s[-1]==str:
        return s[0:-1]
    else:
        return s

def delete_postfix2(s,str):
    if s[0]==‘内‘:
        return s[0:3]
    elif s[-3:]==str:
        return s[0:2]
    else:
        return s
UserInfoData1.UserInfo_19 = UserInfoData1.UserInfo_19.apply(lambda s:delete_postfix2(s,‘自治区‘))
UserInfoData1.UserInfo_19 = UserInfoData1.UserInfo_19.apply(lambda s:delete_postfix1(s,‘省‘))
UserInfoData1.UserInfo_19 = UserInfoData1.UserInfo_19.apply(lambda s:delete_postfix1(s,‘市‘))

import pickle
pickle_file = open(‘dicts.pkl‘, ‘rb‘)
my_list2 = pickle.load(pickle_file)
pickle_file.close()
my_list2.update({‘山东‘:[‘济南‘,‘青岛‘,‘淄博‘,‘枣庄‘,‘东营‘,‘烟台‘,‘潍坊‘,‘济宁‘,‘泰安‘,‘威海‘,‘日照‘,‘莱芜‘,
                       ‘临沂‘,‘德州‘,‘聊城‘,‘滨州‘,‘荷泽‘,‘菏泽‘]})
my_list2.update({‘内蒙古‘:[‘呼和浩特‘,‘包头‘,‘乌海‘,‘赤峰‘,‘通辽‘,‘鄂尔多斯‘,‘呼伦贝尔‘,‘巴彦淖尔‘,‘乌兰察布‘,
                        ‘兴安盟‘,‘锡林郭勒盟‘,‘阿拉善盟‘,‘巴彦淖尔盟‘,‘乌兰察布盟‘]})
my_list2.update({‘贵州‘:[‘贵阳‘,‘六盘水‘,‘遵义‘,‘安顺‘,‘铜仁地区‘,‘黔西南布依族苗族自治州‘,‘毕节地区‘,
                       ‘黔东南苗族侗族自治州‘,‘黔南布依族苗族自治州‘,‘黔南‘,‘黔东南‘,‘毕节‘,‘铜仁‘]})
my_list2.update({‘新疆‘:[‘乌鲁木齐‘,‘克拉玛依‘,‘吐鲁番地区‘,‘哈密地区‘,‘昌吉回族自治州‘,‘博尔塔拉蒙古自治州‘,‘巴音郭楞蒙古自治州‘,‘阿克苏地区‘,‘克孜勒苏柯尔克孜自治州‘,
 ‘喀什地区‘,‘和田地区‘,‘伊犁哈萨克自治州‘,‘塔城地区‘,‘阿勒泰地区‘,‘石河子‘,‘阿拉尔‘,‘图木舒克‘,‘五家渠‘,‘哈密‘,‘阿克苏‘]})

for c,p in zip(UserInfoData1.UserInfo_4,UserInfoData1.UserInfo_7):
    if str(p)==‘nan‘:
        pass
    else:
        province = my_list2.get(p)
        if c in province:
            i = UserInfoData1[UserInfoData1.UserInfo_4==c].index
            UserInfoData1.UserInfo_8[i] = UserInfoData1.UserInfo_4[i]