Merge remote-tracking branch 'origin/master'

alphamind/data/engines/sqlengine.py

@@ -271,8 +271,7 @@ class SqlEngine(object):
                            factors: Union[Transformer, Iterable[object]],
                            start_date: str = None,
                            end_date: str = None,
-                           dates: Iterable[str] = None,
-                           warm_start: int = 0) -> pd.DataFrame:
+                           dates: Iterable[str] = None) -> pd.DataFrame:
 
         if isinstance(factors, Transformer):
             transformer = factors
@@ -282,37 +281,20 @@ class SqlEngine(object):
         dependency = transformer.dependency
         factor_cols = _map_factors(dependency)
 
-        fast_path_optimization = False
-
-        for name in transformer.expressions:
-            if not isinstance(name, SecurityLatestValueHolder) and not isinstance(name, str):
-                break
-        else:
-            fast_path_optimization = True
-
-        if fast_path_optimization:
-            real_start_date = start_date
-            real_end_date = end_date
-            real_dates = dates
-        else:
-            if dates:
-                real_start_date = advanceDateByCalendar('china.sse', dates[0], str(-warm_start) + 'b').strftime(
-                    '%Y-%m-%d')
-                real_end_date = dates[-1]
-            else:
-                real_start_date = advanceDateByCalendar('china.sse', start_date, str(-warm_start) + 'b').strftime(
-                    '%Y-%m-%d')
-                real_end_date = end_date
-            real_dates = None
-
-        cond = universe.query_range(real_start_date, real_end_date, real_dates)
+        cond = universe.query_range(start_date, start_date, dates)
 
         big_table = FullFactorView
 
         for t in set(factor_cols.values()):
             if t.__table__.name != FullFactorView.__table__.name:
-                big_table = outerjoin(big_table, t, and_(FullFactorView.trade_date == t.trade_date,
-                                                         FullFactorView.code == t.code))
+                if dates is not None:
+                    big_table = outerjoin(big_table, t, and_(FullFactorView.trade_date == t.trade_date,
+                                                             FullFactorView.code == t.code,
+                                                             FullFactorView.trade_date.in_(dates)))
+                else:
+                    big_table = outerjoin(big_table, t, and_(FullFactorView.trade_date == t.trade_date,
+                                                             FullFactorView.code == t.code,
+                                                             FullFactorView.trade_date.between(start_date, end_date)))
 
         big_table = join(big_table, UniverseTable,
                          and_(FullFactorView.trade_date == UniverseTable.trade_date,
@@ -328,10 +310,7 @@ class SqlEngine(object):
         for col in res.columns:
             if col not in set(['code', 'isOpen']) and col not in df.columns:
                 df[col] = res[col].values
-        if dates:
-            df = df[df.index.isin(dates)]
-        else:
-            df = df[start_date:end_date]
+
         return df.reset_index()
 
     def fetch_benchmark(self,

alphamind/examples/model_training.py

@@ -29,17 +29,17 @@ re-balance   - 1 week
 training     - every 4 week
 '''
 
-engine = SqlEngine('postgresql+psycopg2://postgres:we083826@localhost/alpha')
-universe = Universe('hs300', ['hs300'])
+engine = SqlEngine('postgresql+psycopg2://postgres:we083826@192.168.0.101/alpha')
+universe = Universe('zz500', ['zz500'])
 neutralize_risk = ['SIZE'] + industry_styles
 portfolio_risk_neutralize = ['SIZE']
 portfolio_industry_neutralize = True
 alpha_factors = ['RVOL', 'EPS', 'CFinc1', 'BDTO', 'VAL', 'CHV', 'GREV', 'ROEDiluted']  # ['BDTO', 'RVOL', 'CHV', 'VAL', 'CFinc1'] # risk_styles
-benchmark = 300
+benchmark = 905
 n_bins = 5
-frequency = '1w'
+frequency = '2w'
 batch = 8
-start_date = '2012-01-01'
+start_date = '2017-01-01'
 end_date = '2017-08-31'
 method = 'risk_neutral'
 use_rank = 100

alphamind/model/data_preparing.py

@@ -11,6 +11,9 @@ from typing import Iterable
 from typing import Union
 from PyFin.api import makeSchedule
 from PyFin.api import BizDayConventions
+from PyFin.api import advanceDateByCalendar
+from PyFin.DateUtilities import Period
+from PyFin.Enums import TimeUnits
 from alphamind.data.transformer import Transformer
 from alphamind.data.engines.sqlengine import SqlEngine
 from alphamind.data.engines.universe import Universe
@@ -19,14 +22,15 @@ from alphamind.utilities import alpha_logger
 
 
 def _map_horizon(frequency: str) -> int:
-    if frequency == '1d':
-        return 0
-    elif frequency == '1w':
-        return 4
-    elif frequency == '1m':
-        return 21
-    elif frequency == '3m':
-        return 62
+    parsed_period = Period(frequency)
+    unit = parsed_period.units()
+    length = parsed_period.length()
+    if unit == TimeUnits.BDays or unit == TimeUnits.Days:
+        return length - 1
+    elif unit == TimeUnits.Weeks:
+        return 5 * length - 1
+    elif unit == TimeUnits.Months:
+        return 22 * length - 1
     else:
         raise ValueError('{0} is an unrecognized frequency rule'.format(frequency))
 
@@ -39,6 +43,9 @@ def prepare_data(engine: SqlEngine,
                  universe: Universe,
                  benchmark: int,
                  warm_start: int = 0):
+    if warm_start > 0:
+        start_date = advanceDateByCalendar('china.sse', start_date, str(-warm_start) + 'b').strftime('%Y-%m-%d')
+
     dates = makeSchedule(start_date, end_date, frequency, calendar='china.sse', dateRule=BizDayConventions.Following)
 
     horizon = _map_horizon(frequency)
@@ -50,8 +57,7 @@ def prepare_data(engine: SqlEngine,
 
     factor_df = engine.fetch_factor_range(universe,
                                           factors=transformer,
-                                          dates=dates,
-                                          warm_start=warm_start).sort_values(['trade_date', 'code'])
+                                          dates=dates).sort_values(['trade_date', 'code'])
     return_df = engine.fetch_dx_return_range(universe, dates=dates, horizon=horizon)
     industry_df = engine.fetch_industry_range(universe, dates=dates)
     benchmark_df = engine.fetch_benchmark_range(benchmark, dates=dates)
@@ -61,7 +67,7 @@ def prepare_data(engine: SqlEngine,
     df = pd.merge(df, industry_df, on=['trade_date', 'code'])
     df['weight'] = df['weight'].fillna(0.)
 
-    return df[['trade_date', 'code', 'dx']], df[
+    return dates, df[['trade_date', 'code', 'dx']], df[
         ['trade_date', 'code', 'weight', 'isOpen', 'industry_code', 'industry'] + transformer.names]
 
 
@@ -140,15 +146,14 @@ def fetch_data_package(engine: SqlEngine,
     alpha_logger.info("Starting data package fetching ...")
 
     transformer = Transformer(alpha_factors)
-    dates = makeSchedule(start_date, end_date, frequency, calendar='china.sse', dateRule=BizDayConventions.Following)
-    return_df, factor_df = prepare_data(engine,
-                                        transformer,
-                                        start_date,
-                                        end_date,
-                                        frequency,
-                                        universe,
-                                        benchmark,
-                                        warm_start)
+    dates, return_df, factor_df = prepare_data(engine,
+                                               transformer,
+                                               start_date,
+                                               end_date,
+                                               frequency,
+                                               universe,
+                                               benchmark,
+                                               warm_start)
 
     if neutralized_risk:
         risk_df = engine.fetch_risk_model_range(universe, dates=dates, risk_model=risk_model)[1]