liuyu/qlib
1
0
Fork 0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-12 07:16:54 +08:00
Code Activity

fix ubg

Browse Source
This commit is contained in:
bxdd
2021-03-29 20:15:42 +08:00
parent 31bc85bf86
commit fb7f84f31e
5 changed files with 17 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/highfreq/highfreq_processor.py
View File

@@ -71,7 +71,7 @@ class HighFreqNorm(Processor):
).sort_index() ).sort_index()
return df_new_features return df_new_features
def config(fit_start_time=None, fit_end_time=None, **kwargs): def config(self, fit_start_time=None, fit_end_time=None, **kwargs):
if fit_start_time: if fit_start_time:
self.fit_start_time = fit_start_time self.fit_start_time = fit_start_time
if fit_end_time: if fit_end_time:

2
examples/highfreq/workflow.py
View File

@@ -31,7 +31,7 @@ class HighfreqWorkflow(object):
SPEC_CONF = {"custom_ops": [DayLast, FFillNan, BFillNan, Date, Select, IsNull, Cut], "expression_cache": None} SPEC_CONF = {"custom_ops": [DayLast, FFillNan, BFillNan, Date, Select, IsNull, Cut], "expression_cache": None}
MARKET = "csi300" MARKET = "all"
start_time = "2020-09-15 00:00:00" start_time = "2020-09-15 00:00:00"
end_time = "2021-01-18 16:00:00" end_time = "2021-01-18 16:00:00"

14
examples/rolling_process_data/workflow.py
View File

@@ -101,15 +101,16 @@ class RollingDataWorkflow(object):
print(f"===========rolling{rolling_offset} start===========") print(f"===========rolling{rolling_offset} start===========")
if rolling_offset: if rolling_offset:
dataset.init( dataset.config(
handler_kwargs={ handler_kwargs={
"init_type": DataHandlerLP.IT_FIT_SEQ,
"start_time": datetime(train_start_time[0] + rolling_offset, *train_start_time[1:]), "start_time": datetime(train_start_time[0] + rolling_offset, *train_start_time[1:]),
"end_time": datetime(test_end_time[0] + rolling_offset, *test_end_time[1:]), "end_time": datetime(test_end_time[0] + rolling_offset, *test_end_time[1:]),
"fit_start_time": datetime(train_start_time[0] + rolling_offset, *train_start_time[1:]), "processor_kwargs":{
"fit_end_time": datetime(train_end_time[0] + rolling_offset, *train_end_time[1:]), "fit_start_time": datetime(train_start_time[0] + rolling_offset, *train_start_time[1:]),
"fit_end_time": datetime(train_end_time[0] + rolling_offset, *train_end_time[1:]),
},
}, },
segment_kwargs={ segments={
"train": ( "train": (
datetime(train_start_time[0] + rolling_offset, *train_start_time[1:]), datetime(train_start_time[0] + rolling_offset, *train_start_time[1:]),
datetime(train_end_time[0] + rolling_offset, *train_end_time[1:]), datetime(train_end_time[0] + rolling_offset, *train_end_time[1:]),
@@ -124,6 +125,9 @@ class RollingDataWorkflow(object):
), ),
}, },
) )
dataset.setup_data(
handler_kwargs={"init_type": DataHandlerLP.IT_FIT_SEQ,}
)
dtrain, dvalid, dtest = dataset.prepare(["train", "valid", "test"]) dtrain, dvalid, dtest = dataset.prepare(["train", "valid", "test"])
print(dtrain, dvalid, dtest) print(dtrain, dvalid, dtest)

4
qlib/data/dataset/handler.py
View File

@@ -407,7 +407,7 @@ class DataHandlerLP(DataHandler):
if self.drop_raw: if self.drop_raw:
del self._data del self._data
def config(self, processors_kwargs:dict = None, **kwargs): def config(self, processor_kwargs:dict = None, **kwargs):
""" """
configuration of data. configuration of data.
# what data to be loaded from data source # what data to be loaded from data source
@@ -417,7 +417,7 @@ class DataHandlerLP(DataHandler):
""" """
super().config(**kwargs) super().config(**kwargs)
if processors_kwargs is not None: if processor_kwargs is not None:
for processor in self.get_all_processors(): for processor in self.get_all_processors():
processor.config(**processor_kwargs) processor.config(**processor_kwargs)

8
qlib/data/dataset/processor.py
View File

@@ -72,7 +72,7 @@ class Processor(Serializable):
""" """
return True return True
def config(**kwargs): def config(self, **kwargs):
super().config(kwargs.get("dump_all", None), kwargs.get("exclude", None)) super().config(kwargs.get("dump_all", None), kwargs.get("exclude", None))
@@ -195,7 +195,7 @@ class MinMaxNorm(Processor):
df.loc(axis=1)[self.cols] = normalize(df[self.cols].values) df.loc(axis=1)[self.cols] = normalize(df[self.cols].values)
return df return df
def config(fit_start_time=None, fit_end_time=None, **kwargs): def config(self, fit_start_time=None, fit_end_time=None, **kwargs):
if fit_start_time: if fit_start_time:
self.fit_start_time = fit_start_time self.fit_start_time = fit_start_time
if fit_end_time: if fit_end_time:
@@ -230,7 +230,7 @@ class ZScoreNorm(Processor):
df.loc(axis=1)[self.cols] = normalize(df[self.cols].values) df.loc(axis=1)[self.cols] = normalize(df[self.cols].values)
return df return df
def config(fit_start_time=None, fit_end_time=None, **kwargs): def config(self, fit_start_time=None, fit_end_time=None, **kwargs):
if fit_start_time: if fit_start_time:
self.fit_start_time = fit_start_time self.fit_start_time = fit_start_time
if fit_end_time: if fit_end_time:
@@ -273,7 +273,7 @@ class RobustZScoreNorm(Processor):
df.clip(-3, 3, inplace=True) df.clip(-3, 3, inplace=True)
return df return df
def config(fit_start_time=None, fit_end_time=None, **kwargs): def config(self, fit_start_time=None, fit_end_time=None, **kwargs):
if fit_start_time: if fit_start_time:
self.fit_start_time = fit_start_time self.fit_start_time = fit_start_time
if fit_end_time: if fit_end_time: