def __init__(self, region_query=None):
# 设置查询结果
if region_query is None:
mongo = MongoDB()
mdb = MonDatabase(mongodb=mongo, database_name='region')
collection = MonCollection(database=mdb, collection_name='admincode')
self.collection = collection.collection
else:
self.collection = region_query
self.collection = None
def search_from_dbase(variables=None, query_dict=None, match='exact'):
collection_variable = MonCollection(database=MonDatabase(
mongodb=MongoDB(), database_name='region'),
collection_name='storedvariable')
found = collection_variable.find(query_dict)
found_dict = {item['origin']: item['variable'] for item in found}
if match == 'exact':
pd_result = VariableMatcher.search_for_same_variable(
variables=variables, source=found_dict.keys())
for ind in pd_result.index:
pd_result.loc[ind, 'matched_variable'] = found_dict.get(
pd_result.loc[ind, 'matched_variable'])
else:
pd_result = VariableMatcher.search_for_similar_variable(
variables=variables, source=found_dict.keys())
for ind in pd_result.index:
pd_result.loc[ind, 'matched_middel_variable'] = pd_result.loc[
ind, 'matched_variable']
pd_result.loc[ind, 'matched_variable'] = found_dict.get(
pd_result.loc[ind, 'matched_variable'])
return pd_result
return ref_regions_dict
@property
def matched_region(self):
return self._result
if __name__ == '__main__':
pop_year = '2010'
pop_region_file_2010 = r'E:\data\popcensus\origin\var_temp.xls'
raw_region_2010 = Excel(pop_region_file_2010).read()
to_be_matched = [re.sub('\s+','',item[0]) for item in raw_region_2010 if re.match('^\s*$',item[0]) is None]
pd_to_be_matched = pd.DataFrame(to_be_matched,columns=['region'])
pd_to_be_matched['rid'] = range(pd_to_be_matched.shape[0])
collection = MonCollection(database=MonDatabase(mongodb=MongoDB(), database_name='region'), collection_name='admincode')
found = collection.collection.find(filter={'year':'2010'},
projection={'acode':True,'region':True,'_id':True},
sort=[('acode',1)])
pd_to_be_compared = pd.DataFrame(list(found))
pd_to_be_compared['cid'] = range(pd_to_be_compared.shape[0])
#pd_to_be_compared['_id'] = pd_to_be_compared['_id'].apply(str)
print(pd_to_be_matched,pd_to_be_compared)
algo = RegionMatchingOrderAlgorithm(pd_to_be_matched,pd_to_be_compared)
# 首先是寻找可靠的匹配作为锚点
algo.find_anchor()
# 其次进行顺序的严格匹配
algo.exactly_matching_from_region_set()
print(algo.matched_region)
# coding = UTF-8
from libs.imexport.class_mongodb import MongoDB, MonDatabase, MonCollection
# 0. 连接数据库
collection_variable = MonCollection(database=MonDatabase(
mongodb=MongoDB(), database_name='variable'),
collection_name='referencevariable')
# 1. 参数设置
CEIC_VARIABLE = False
CHINASTAT_VARIABLE = True
# 2. 导入CEIC变量
if CEIC_VARIABLE:
ceic_collection = MonCollection(database=MonDatabase(
mongodb=MongoDB(), database_name='region'),
collection_name='ceic')
refer_variables = ceic_collection.collection.find().distinct('variable')
source = 'CEIC'
# 3. 导入中国统计年鉴变量
if CHINASTAT_VARIABLE:
Chinastat_collection = MonCollection(database=MonDatabase(
mongodb=MongoDB(), database_name='region'),
collection_name='provincestat')
refer_variables = Chinastat_collection.collection.find().distinct(
'variable')
source = '中国统计年鉴'
if isinstance(refer_variables, list):
def __init__(self):
# 连接AdminDatabase集合
mongo = MongoDB()
mdb = MonDatabase(mongodb=mongo, database_name='region')
self.collection = MonCollection(database=mdb,
collection_name='admincode')
USER_VARIABLES = None
USER_PERIOD = None
USER_REGION = None
USER_FILTER = dict()
USER_PROJECTION = {'_id': 0, 'variable': 1, 'value': 1, 'acode': 1, 'year': 1}
USER_SORT = [('acode', 1), ('year', 1)]
# 平滑面板时固定的轴名称
USER_FIXED_INDEX = 'acode'
# 数据集最小数量
USER_MIN_DATASET_NUMBER = 50
# 1. to query in the mongodb
# 1.1 连接数据库
user_database = MonDatabase(mongodb=MongoDB(),
database_name=DATABASES.get(DATABASE_CHOICE))
user_colllection = COLLECTIONS.get(COLLECTION_CHOICE)(database=user_database)
# 1.2 数据库基本参数打印
if COLLECTION_INFO:
user_colllection.info()
round = 1
while True:
# 1.3 查询数据，返回结果
# 打印开始信息
print(''.join(['=' * 40, 'Round ', str(round), '=' * 40]))
# 设定变量variables
if VARIABLE_RANDOM_CHOICE > 0:
variables = random.sample(user_colllection.variables,
def __init__(self):
# 连接PopCensus集合
mongo = MongoDB()
mdb = MonDatabase(mongodb=mongo, database_name='region')
self.collection = MonCollection(database=mdb, collection_name='popcensus')
# coding = UTF-8
import re
from libs.database.class_mondbprovincestat import MonDBProvinceStat
from libs.imexport.class_mongodb import MongoDB, MonDatabase, MonCollection
mongo = MongoDB()
mdb = MonDatabase(mongodb=mongo, database_name='region')
prostat = MonDBProvinceStat(database=mdb)
# 1. 修正变量名重合的问题
variables_origin = set(prostat.variables)
variables_no_space = set([re.sub('\s+', '', var) for var in prostat.variables])
i = 1
for item in sorted(prostat.variables):
print(i, ': ', item, ' -- ', len(item))
i += 1
:param fill_value: 详见pandas.pivot_table()函数参数说明
:return: 返回转换后的宽格式表格
:rtype: pandas.DataFrame
"""
result = pd.pivot_table(data=dataframe,
values=values,
index=index,
columns=columns,
dropna=dropna,
fill_value=fill_value)
return result
if __name__ == '__main__':
mcollection = MonCollection(database=MonDatabase(mongodb=MongoDB(),
database_name='region'),
collection_name='provincestat')
cursor = mcollection.find(
{
'variable': {
'$in': ['人均地区生产总值', '私人控股企业法人单位数', '城镇居民消费', '城镇单位就业人员平均工资']
}
},
projection={
'_id': 0,
'variable': 1,
'value': 1,
'province': 1,
'acode': 1,
'year': 1
})
def __init__(self,
database=MonDatabase(mongodb=MongoDB(),
database_name='region'),
collection_name='ceic'):
super().__init__(database=database, collection_name=collection_name)
def setUp(self):
mongo = MongoDB(
conn_str='mongodb://*****:*****@139.196.189.191:3717/')
mdb = MonDatabase(mongodb=mongo, database_name='region')
self.mcollection = MonCollection(database=mdb,
collection_name='cities')