liuyu/qlib
1
0
Fork 0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-16 09:11:00 +08:00
Code Activity

black

Browse Source
This commit is contained in:
wangershi
2021-03-07 19:35:50 +08:00
parent 6bcd88973b
commit 9df0361262
1 changed files with 13 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
scripts/data_collector/utils.py
View File

@@ -105,7 +105,11 @@ def return_date_list(source_dir, date_field_name: str, file_path: Path):
def get_calendar_list_by_ratio( def get_calendar_list_by_ratio(
source_dir: [str, Path], date_field_name: str = "date", threshold: float = 0.5, minimum_count: int = 10, max_workers: int = 16 source_dir: [str, Path],
date_field_name: str = "date",
threshold: float = 0.5,
minimum_count: int = 10,
max_workers: int = 16,
) -> list: ) -> list:
"""get calendar list by selecting the date when few funds trade in this day """get calendar list by selecting the date when few funds trade in this day
@@ -134,7 +138,7 @@ def get_calendar_list_by_ratio(
_number_all_funds = len(file_list) _number_all_funds = len(file_list)
logger.info(f"count how many funds trade in this day......") logger.info(f"count how many funds trade in this day......")
_dict_count_trade = dict() # dict{date:count} _dict_count_trade = dict() # dict{date:count}
_fun = partial(return_date_list, source_dir, date_field_name) _fun = partial(return_date_list, source_dir, date_field_name)
with tqdm(total=_number_all_funds) as p_bar: with tqdm(total=_number_all_funds) as p_bar:
with ProcessPoolExecutor(max_workers=max_workers) as executor: with ProcessPoolExecutor(max_workers=max_workers) as executor:
@@ -148,7 +152,7 @@ def get_calendar_list_by_ratio(
p_bar.update() p_bar.update()
logger.info(f"count how many funds have founded in this day......") logger.info(f"count how many funds have founded in this day......")
_dict_count_founding = {date:_number_all_funds for date in _dict_count_trade.keys()} # dict{date:count} _dict_count_founding = {date: _number_all_funds for date in _dict_count_trade.keys()} # dict{date:count}
with tqdm(total=_number_all_funds) as p_bar: with tqdm(total=_number_all_funds) as p_bar:
with ProcessPoolExecutor(max_workers=max_workers) as executor: with ProcessPoolExecutor(max_workers=max_workers) as executor:
for date_list in executor.map(_fun, file_list[:_number_all_funds]): for date_list in executor.map(_fun, file_list[:_number_all_funds]):
@@ -157,7 +161,11 @@ def get_calendar_list_by_ratio(
if date < oldest_date: if date < oldest_date:
_dict_count_founding[date] -= 1 _dict_count_founding[date] -= 1
calendar = [date for date in _dict_count_trade if _dict_count_trade[date] >= max(int(_dict_count_founding[date] * threshold), minimum_count)] calendar = [
date
for date in _dict_count_trade
if _dict_count_trade[date] >= max(int(_dict_count_founding[date] * threshold), minimum_count)
]
return calendar return calendar