def run(self):
if self.must_factors:
musts = [
factor.agg("and", axis="columns")
for factor in self.must_factors
]
self.must_result = list(accumulate(musts,
func=operator.__and__))[-1]
if self.score_factors:
scores = [
factor.agg("mean", axis="columns")
for factor in self.score_factors
]
self.score_result = list(accumulate(scores,
func=operator.__add__))[-1]
if self.must_result:
self.score_result = self.score_result[
self.must_result > 0 and self.score_result > self.threshold]
else:
self.score_result = self.score_result[
self.score_result > self.threshold]
self.score_result.name = 'score'
self.df = self.score_result.reset_index()
self.df = index_df_with_time(self.df)
print(self.df)
def get_data(data_schema,
security_id=None,
codes=None,
level=None,
provider='eastmoney',
columns=None,
return_type='df',
start_timestamp=None,
end_timestamp=None,
filters=None,
session=None,
order=None,
limit=None):
local_session = False
if not session:
store_category = get_store_category(data_schema)
session = get_db_session(provider=provider,
store_category=store_category)
local_session = True
try:
if columns:
query = session.query(*columns)
else:
query = session.query(data_schema)
if security_id:
query = query.filter(data_schema.security_id == security_id)
if codes:
query = query.filter(data_schema.code.in_(codes))
if level:
query = query.filter(data_schema.level == level)
query = common_filter(query,
data_schema=data_schema,
start_timestamp=start_timestamp,
end_timestamp=end_timestamp,
filters=filters,
order=order,
limit=limit)
if return_type == 'df':
df = pd.read_sql(query.statement, query.session.bind)
if not df.empty:
return index_df_with_time(df, drop=False)
elif return_type == 'domain':
return query.all()
elif return_type == 'dict':
return [item.to_json() for item in query.all()]
except Exception:
raise
finally:
if local_session:
session.close()
def run(self):
if self.must_factors:
musts = []
for factor in self.must_factors:
df = factor.get_result_df()
if len(df.columns) > 1:
s = df.agg("and", axis="columns")
s.name = 'score'
musts.append(s.to_frame(name='score'))
else:
df.columns = ['score']
musts.append(df)
self.must_result = list(accumulate(musts,
func=operator.__and__))[-1]
if self.score_factors:
scores = []
for factor in self.score_factors:
df = factor.get_result_df()
if len(df.columns) > 1:
s = df.agg("mean", axis="columns")
s.name = 'score'
scores.append(s.to_frame(name='score'))
else:
df.columns = ['score']
scores.append(df)
self.score_result = list(accumulate(scores,
func=operator.__add__))[-1]
if (self.must_result is not None) and (self.score_result is not None):
result = self.score_result[self.must_result.score and (
self.score_result.score >= self.threshold)]
elif self.score_result is not None:
result = self.score_result[
self.score_result.score >= self.threshold]
else:
result = self.must_result[self.must_result.score]
self.result_df = result.reset_index()
self.result_df = index_df_with_time(self.result_df)
def run(self):
if self.must_factors:
musts = []
for factor in self.must_factors:
df = factor.get_df()
if len(df.columns) > 1:
musts.append(df.agg("and", axis="columns"))
else:
musts.append(df)
self.must_result = list(accumulate(musts,
func=operator.__and__))[-1]
if self.score_factors:
scores = []
for factor in self.score_factors:
df = factor.get_df()
if len(df.columns) > 1:
scores.append(df.agg("mean", axis="columns"))
else:
scores.append(df)
self.score_result = list(accumulate(scores,
func=operator.__add__))[-1]
if (self.must_result is not None) and (self.score_result is not None):
self.result = self.score_result[
self.must_result > 0 and self.score_result > self.threshold]
elif self.score_result is not None:
self.result = self.score_result[self.score_result > self.threshold]
else:
self.result = self.must_result
self.result.name = 'score'
self.df = self.result.reset_index()
self.df = index_df_with_time(self.df)
print(self.df)
def init_history_data(self, history_data):
self.history_data = pd.DataFrame(history_data)
self.history_data = index_df_with_time(self.history_data)
self.current_timestamp = to_pd_timestamp(history_data[-1]['timestamp'])
self.current_data = history_data[-1]
def get_data(data_schema,
security_list=None,
security_id=None,
codes=None,
level=None,
provider='eastmoney',
columns=None,
return_type='df',
start_timestamp=None,
end_timestamp=None,
filters=None,
session=None,
order=None,
limit=None):
local_session = False
if not session:
store_category = get_store_category(data_schema)
session = get_db_session(provider=provider,
store_category=store_category)
local_session = True
try:
if columns:
if data_schema.timestamp not in columns:
columns.append(data_schema.timestamp)
query = session.query(*columns)
else:
query = session.query(data_schema)
if security_id:
query = query.filter(data_schema.security_id == security_id)
if codes:
query = query.filter(data_schema.code.in_(codes))
if security_list:
query = query.filter(data_schema.security_id.in_(security_list))
# we always store different level in different schema,the level param is not useful now
if level:
try:
# some schema has no level,just ignore it
data_schema.level
if type(level) == TradingLevel:
level = level.value
query = query.filter(data_schema.level == level)
except Exception as e:
pass
query = common_filter(query,
data_schema=data_schema,
start_timestamp=start_timestamp,
end_timestamp=end_timestamp,
filters=filters,
order=order,
limit=limit)
if return_type == 'df':
df = pd.read_sql(query.statement, query.session.bind)
if not df.empty:
return index_df_with_time(df, drop=False)
elif return_type == 'domain':
return query.all()
elif return_type == 'dict':
return [item.to_json() for item in query.all()]
except Exception:
raise
finally:
if local_session:
session.close()
