liuyu/qlib
1
0
Fork 0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-15 00:36:55 +08:00
Code Activity

Add a check if change is mutated to YahooNormalize1d

Browse Source
This commit is contained in:
zhupr
2021-07-15 22:12:53 +08:00
committed by you-n-g
parent 2023f714c9
commit c97a96363d
1 changed files with 16 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
scripts/data_collector/yahoo/collector.py
View File

@@ -325,9 +325,22 @@ class YahooNormalize(BaseNormalize):
# NOTE: The data obtained by Yahoo finance sometimes has exceptions # NOTE: The data obtained by Yahoo finance sometimes has exceptions
# WARNING: If it is normal for a `symbol(exchange)` to differ by a factor of *89* to *111* for consecutive trading days, # WARNING: If it is normal for a `symbol(exchange)` to differ by a factor of *89* to *111* for consecutive trading days,
# WARNING: the logic in the following line needs to be modified # WARNING: the logic in the following line needs to be modified
_mask = (change_series >= 89) & (change_series <= 111) _count = 0
_tmp_cols = ["high", "close", "low", "open", "adjclose"] while True:
df.loc[_mask, _tmp_cols] = df.loc[_mask, _tmp_cols] / 100 # NOTE: may appear unusual for many days in a row
change_series = YahooNormalize.calc_change(df, last_close)
_mask = (change_series >= 89) & (change_series <= 111)
if not _mask.any():
break
_tmp_cols = ["high", "close", "low", "open", "adjclose"]
df.loc[_mask, _tmp_cols] = df.loc[_mask, _tmp_cols] / 100
_count += 1
if _count >= 10:
_symbol = df.loc[df[symbol_field_name].first_valid_index()]["symbol"]
logger.warning(
f"{_symbol} `change` is abnormal for {_count} consecutive days, please check the specific data file carefully"
)
df["change"] = YahooNormalize.calc_change(df, last_close) df["change"] = YahooNormalize.calc_change(df, last_close)
columns += ["change"] columns += ["change"]