import numpy as np
a=np.load('ip_explain_by_geoip2_china.npy',allow_pickle=True)
ip_exp=a.item()
temp = pd.DataFrame(list(ip_exp.items()), columns=['ip', 'ip_exp'])
temp[['country','province_exp','city_exp']] = temp['ip_exp'].str.split('_', expand = True)
res = data[['ip','province','city','label']]
temp = temp.merge(res, on = 'ip', how = 'left')
temp['province'] = temp['province_exp'].apply(lambda x:x[:2] if x != None else '-1')
temp['province_final'] = temp['province']
temp['city'][temp.city.isnull()] = temp['city_exp'][temp.city.isnull()]
temp['city'] = temp['city'].apply(lambda x:str(x)[:3] if x != None else '-1')
temp['city_final'] = temp['city']
temp = temp[['ip','country','province_final','city_final']]
ta = data.merge(temp, how = 'left', on = 'ip')
ip处理
最新推荐文章于 2022-10-11 16:52:03 发布