liuyu/qlib
1
0
Fork 0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-10 14:26:56 +08:00
Code Activity

Merge nested main (#597)

Browse Source 
* MVP for Indian Stocks in qlib using yahooquery

* cleaned with black

* cleaned with black

* add YahooNormalizeIN and YahooNormalizeIN1d

* cleaned the code

* added 1min for IN and also updated readme

* update comments

* fix comments

* recorder support upload both raw file and directory

* fix comments

* Update README.md

* Fix docs of QlibRecorder

* sort index after loader (#538)

make sure the fetch method is based on a index-sorted pd.DataFrame

* refactor online serving rolling api

* refactor TRA

* format by black

* fix horizon

* fix TRA when use single head

* clean up

* improve pretrain

* update README

* fix tra when logdir is None

* fix tra when logdir is None

* Update strategy.py

* Update README.md

* Update README.md

* Conda Suggestion

* code standard docs

* Update ensemble.py (#560)

* Fix CI  Bug (#575)


Co-authored-by: yuxwang <anduinnn@foxmail.com>

* Update gen.py (#576)

* Fix multi-process loop calls (#574)

* check lexsort in the 'lazy_sort_index' function (#566)

* check lexsort

* check lexsort

* lexsort comment

* lexsort comment

* Delete .DS_Store

* Update README.md

* bug fix & use oracle transport pretrain

* mend

* Add `backend_freq_config` parameter, support multi-freq uri

* Add sample_config to QlibDataLoader, support multi-freq

* add multi-freq example

* get_cls_kwargs renamed get_callable_kwargs

* support multi-freq uri

* Add inst_processors to D.features

* Fix typo

* Fix the index type of the multi-freq example

* Fix duplicate mlflow directories in tests

* Add DataPathManager to QlibConfig && modify inst_processors to supports list only

* Modify the default value in the multi_freq example

* Modify client-server mode and dataset-cache to disable inst_processor

* Add wheel package to github CI

* fix comment

* Update FAQ.rst

* Update README.md

Fix wrong link

* Update the docs of TaskManager (#586)

* Update manage.py

* update yaml

* update run_all_model

* Modify the Feature to be case sensitive (#589)

* update README

* remove verbose

* fix spell bug

* fix typos (#592)

* Update Release Note

* fix portfolio bug

* Add calendar support for resample

* add freq kwargs

* test.yml: Remove redundant code (#595)

* Supporting shared processor (#596)

* Supporting shared processor

* fix readonly reverse bug

* remove pytests dependency

* with fit bug

* fix parameter error

* fix comments

* Fix undefined names in Python code (#599)

* Update pytorch_tabnet.py

$ `flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics`
```
./qlib/qlib/contrib/model/pytorch_tabnet.py:567:38: F821 undefined name 'inp'
            self.independ.append(GLU(inp, out_dim, vbs=vbs))
                                     ^
./qlib/examples/model_rolling/task_manager_rolling.py:75:18: F821 undefined name 'task_train'
        run_task(task_train, self.task_pool, experiment_name=self.experiment_name)
                 ^
2     F821 undefined name 'task_train'
2
```

* Fix undefined names in Python code

* from qlib.model.trainer import task_train

* update seed

* fix some docstring

* add comments

* Fix SimpleDatasetCache

* Update setup.py

updated classifiers

* Update setup.py

change to matplotlib==3.3

* Update python-publish.yml

added python 3.9

* updategrade version number

* Update model list

* fix the type of filter_pipe

* fix comment

* fix record_temp

* update cvxpy version

* Update code_standard.rst (#587)

* Update code_standard.rst

* Update docs/developer/code_standard.rst

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

* Add file lock for MLflowExpManager (#619)

* fix torch version

* Share version number (#620)

* Update initialization.rst (#622)

* Update initialization.rst

* Update docs/start/initialization.rst

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

* Update docs/start/initialization.rst

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

* fix bugs for running previous exmaple

* fix deal amount bug

* update change doc (#623)

* Add files via upload

* Update README.md

* Update README.md

* Update README.md

* Delete change doc.gif

* Add files via upload

* Update README.md

* Delete change doc.gif

* Add files via upload

* Delete change doc.gif

* Add files via upload

* Update README.md

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

* update doc

* simplify run all model

* fix run all model bug

* Fix Models (#483)

* fix gat dataset

* fix tft model

* Update tft.py

* Fix tft.py

Co-authored-by: Pengrong Zhu <zhu.pengrong@foxmail.com>

* type and skip empty exp

* fix model yaml config

* fix tft import bug

* skip empty result

* fix model and yaml bug

* fix wrong generate parameter

* Modify multi-freq example (#626)

* modify the example of multi-freq

* add Copyright

* add a comment to average_ops.py

* modify the example of multi-freq

* add comment to multi_freq_handler.py

* add the Ref expression description to multi_freq_handler.py

* add expression description to multi_freq_handler.py

* update images

* fix workflow and update framework

Co-authored-by: Gaurav <2796gaurav@gmail.com>
Co-authored-by: 2796gaurav <17353992+2796gaurav@users.noreply.github.com>
Co-authored-by: bxdd <bxd98@126.com>
Co-authored-by: Young <afe.young@gmail.com>
Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>
Co-authored-by: Dong Zhou <Zhou.Dong@microsoft.com>
Co-authored-by: ZhangTP1996 <ztp18@mails.tsinghua.edu.cn>
Co-authored-by: demon143 <59681577+demon143@users.noreply.github.com>
Co-authored-by: Wangwuyi123 <51237097+Wangwuyi123@users.noreply.github.com>
Co-authored-by: yuxwang <anduinnn@foxmail.com>
Co-authored-by: Pengrong Zhu <zhu.pengrong@foxmail.com>
Co-authored-by: Mark Zhao <50850474+markzhao98@users.noreply.github.com>
Co-authored-by: cslwqxx <cslwqxx@users.noreply.github.com>
Co-authored-by: Dong Zhou <evanzd@users.noreply.github.com>
Co-authored-by: SaintMalik <37118134+saintmalik@users.noreply.github.com>
Co-authored-by: Christian Clauss <cclauss@me.com>
Co-authored-by: Anurag Kumar <mailanu98@gmail.com>
Co-authored-by: demon143 <785696300@qq.com>
This commit is contained in:
wangwenxi-handsome
2021-10-01 02:15:30 +08:00
committed by GitHub
parent 163e3c6266
commit 3760a18a8d
145 changed files with 3982 additions and 1221 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/benchmarks/ALSTM/workflow_config_alstm_Alpha158.yaml
View File

@@ -93,8 +93,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/ALSTM/workflow_config_alstm_Alpha360.yaml
View File

@@ -83,8 +83,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/CatBoost/workflow_config_catboost_Alpha158.yaml
View File

@@ -65,8 +65,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/CatBoost/workflow_config_catboost_Alpha360.yaml
View File

@@ -72,8 +72,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_Alpha158.yaml
View File

@@ -90,8 +90,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_Alpha360.yaml
View File

@@ -97,8 +97,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/GATs/workflow_config_gats_Alpha158.yaml
View File

@@ -91,8 +91,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/GATs/workflow_config_gats_Alpha360.yaml
View File

@@ -83,8 +83,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/GRU/workflow_config_gru_Alpha158.yaml
View File

@@ -92,8 +92,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/GRU/workflow_config_gru_Alpha360.yaml
View File

@@ -82,8 +82,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/LSTM/workflow_config_lstm_Alpha158.yaml
View File

@@ -92,8 +92,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/LSTM/workflow_config_lstm_Alpha360.yaml
View File

@@ -82,8 +82,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

18
examples/benchmarks/LightGBM/features_resample_N.py Normal file
View File

@@ -0,0 +1,18 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
import pandas as pd
from qlib.data.inst_processor import InstProcessor
from qlib.utils.resam import resam_calendar
class ResampleNProcessor(InstProcessor):
def __init__(self, target_frq: str, **kwargs):
self.target_frq = target_frq
def __call__(self, df: pd.DataFrame, *args, **kwargs):
df.index = pd.to_datetime(df.index)
res_index = resam_calendar(df.index, "1min", self.target_frq)
df = df.resample(self.target_frq).last().reindex(res_index)
return df

135
examples/benchmarks/LightGBM/multi_freq_handler.py Normal file
View File

@@ -0,0 +1,135 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
import pandas as pd
from qlib.data.dataset.loader import QlibDataLoader
from qlib.contrib.data.handler import DataHandlerLP, _DEFAULT_LEARN_PROCESSORS, check_transform_proc
class Avg15minLoader(QlibDataLoader):
def load(self, instruments=None, start_time=None, end_time=None) -> pd.DataFrame:
df = super(Avg15minLoader, self).load(instruments, start_time, end_time)
if self.is_group:
# feature_day(day freq) and feature_15min(1min freq, Average every 15 minutes) renamed feature
df.columns = df.columns.map(lambda x: ("feature", x[1]) if x[0].startswith("feature") else x)
return df
class Avg15minHandler(DataHandlerLP):
def __init__(
self,
instruments="csi500",
start_time=None,
end_time=None,
freq="day",
infer_processors=[],
learn_processors=_DEFAULT_LEARN_PROCESSORS,
fit_start_time=None,
fit_end_time=None,
process_type=DataHandlerLP.PTYPE_A,
filter_pipe=None,
inst_processor=None,
**kwargs,
):
infer_processors = check_transform_proc(infer_processors, fit_start_time, fit_end_time)
learn_processors = check_transform_proc(learn_processors, fit_start_time, fit_end_time)
data_loader = Avg15minLoader(
config=self.loader_config(), filter_pipe=filter_pipe, freq=freq, inst_processor=inst_processor
)
super().__init__(
instruments=instruments,
start_time=start_time,
end_time=end_time,
data_loader=data_loader,
infer_processors=infer_processors,
learn_processors=learn_processors,
process_type=process_type,
)
def loader_config(self):
# Results for dataset: df: pd.DataFrame
# len(df.columns) == 6 + 6 * 16, len(df.index.get_level_values(level="datetime").unique()) == T
# df.columns: close0, close1, ..., close16, open0, ..., open16, ..., vwap16
# freq == day:
# close0, open0, low0, high0, volume0, vwap0
# freq == 1min:
# close1, ..., close16, ..., vwap1, ..., vwap16
# df.index.name == ["datetime", "instrument"]: pd.MultiIndex
# Example:
# feature ... label
# close0 open0 low0 ... vwap1 vwap16 LABEL0
# datetime instrument ...
# 2020-10-09 SH600000 11.794546 11.819587 11.769505 ... NaN NaN -0.005214
# 2020-10-15 SH600000 12.044961 11.944795 11.932274 ... NaN NaN -0.007202
# ... ... ... ... ... ... ... ...
# 2021-05-28 SZ300676 6.369684 6.495406 6.306568 ... NaN NaN -0.001321
# 2021-05-31 SZ300676 6.601626 6.465643 6.465130 ... NaN NaN -0.023428
# features day: len(columns) == 6, freq = day
# $close is the closing price of the current trading day:
# if the user needs to get the `close` before the last T days, use Ref($close, T-1), for example:
# $close Ref($close, 1) Ref($close, 2) Ref($close, 3) Ref($close, 4)
# instrument datetime
# SH600519 2021-06-01 244.271530
# 2021-06-02 242.205917 244.271530
# 2021-06-03 242.229889 242.205917 244.271530
# 2021-06-04 245.421524 242.229889 242.205917 244.271530
# 2021-06-07 247.547089 245.421524 242.229889 242.205917 244.271530
# WARNING: Ref($close, N), if N == 0, Ref($close, N) ==> $close
fields = ["$close", "$open", "$low", "$high", "$volume", "$vwap"]
# names: close0, open0, ..., vwap0
names = list(map(lambda x: x.strip("$") + "0", fields))
config = {"feature_day": (fields, names)}
# features 15min: len(columns) == 6 * 16, freq = 1min
# $close is the closing price of the current trading day:
# if the user gets 'close' for the i-th 15min of the last T days, use `Ref(Mean($close, 15), (T-1) * 240 + i * 15)`, for example:
# Ref(Mean($close, 15), 225) Ref(Mean($close, 15), 465) Ref(Mean($close, 15), 705)
# instrument datetime
# SH600519 2021-05-31 241.769897 243.077942 244.712997
# 2021-06-01 244.271530 241.769897 243.077942
# 2021-06-02 242.205917 244.271530 241.769897
# WARNING: Ref(Mean($close, 15), N), if N == 0, Ref(Mean($close, 15), N) ==> Mean($close, 15)
# Results of the current script:
# time: 09:00 --> 09:14, ..., 14:45 --> 14:59
# fields: Ref(Mean($close, 15), 225), ..., Mean($close, 15)
# name: close1, ..., close16
#
# Expression description: take close as an example
# Mean($close, 15) ==> df["$close"].rolling(15, min_periods=1).mean()
# Ref(Mean($close, 15), 15) ==> df["$close"].rolling(15, min_periods=1).mean().shift(15)
# NOTE: The last data of each trading day, which is the average of the i-th 15 minutes
# Average:
# Average of the i-th 15-minute period of each trading day: 1 <= i <= 250 // 16
# Avg(15minutes): Ref(Mean($close, 15), 240 - i * 15)
#
# Average of the first 15 minutes of each trading day; i = 1
# Avg(09:00 --> 09:14), df.index.loc["09:14"]: Ref(Mean($close, 15), 240- 1 * 15) ==> Ref(Mean($close, 15), 225)
# Average of the last 15 minutes of each trading day; i = 16
# Avg(14:45 --> 14:59), df.index.loc["14:59"]: Ref(Mean($close, 15), 240 - 16 * 15) ==> Ref(Mean($close, 15), 0) ==> Mean($close, 15)
# 15min resample to day
# df.resample("1d").last()
tmp_fields = []
tmp_names = []
for i, _f in enumerate(fields):
_fields = [f"Ref(Mean({_f}, 15), {j * 15})" for j in range(1, 240 // 15)]
_names = [f"{names[i][:-1]}{int(names[i][-1])+j}" for j in range(240 // 15 - 1, 0, -1)]
_fields.append(f"Mean({_f}, 15)")
_names.append(f"{names[i][:-1]}{int(names[i][-1])+240 // 15}")
tmp_fields += _fields
tmp_names += _names
config["feature_15min"] = (tmp_fields, tmp_names)
# label
config["label"] = (["Ref($close, -2)/Ref($close, -1) - 1"], ["LABEL0"])
return config

2
examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml
View File

@@ -66,8 +66,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha360.yaml
View File

@@ -73,8 +73,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

4
examples/benchmarks/LightGBM/workflow_config_lightgbm_configurable_dataset.yaml
View File

@@ -81,9 +81,7 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config
config: *port_analysis_config

86
examples/benchmarks/LightGBM/workflow_config_lightgbm_multi_freq.yaml Normal file
View File

@@ -0,0 +1,86 @@
qlib_init:
provider_uri:
day: "~/.qlib/qlib_data/cn_data"
1min: "~/.qlib/qlib_data/cn_data_1min"
region: cn
dataset_cache: null
maxtasksperchild: null
market: &market csi300
benchmark: &benchmark SH000300
data_handler_config: &data_handler_config
start_time: 2008-01-01
# 1min closing time is 15:00:00
end_time: "2020-08-01 15:00:00"
fit_start_time: 2008-01-01
fit_end_time: 2014-12-31
instruments: *market
freq:
label: day
feature_15min: 1min
feature_day: day
# with label as reference
inst_processor:
feature_15min:
- class: ResampleNProcessor
module_path: features_resample_N.py
kwargs:
target_frq: 1d
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
limit_threshold: 0.095
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: LGBModel
module_path: qlib.contrib.model.gbdt
kwargs:
loss: mse
colsample_bytree: 0.8879
learning_rate: 0.2
subsample: 0.8789
lambda_l1: 205.6999
lambda_l2: 580.9768
max_depth: 8
num_leaves: 210
num_threads: 20
dataset:
class: DatasetH
module_path: qlib.data.dataset
kwargs:
handler:
class: Avg15minHandler
module_path: multi_freq_handler.py
kwargs: *data_handler_config
segments:
train: [2008-01-01, 2014-12-31]
valid: [2015-01-01, 2016-12-31]
test: [2017-01-01, 2020-08-01]
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config

2
examples/benchmarks/Linear/workflow_config_linear_Alpha158.yaml
View File

@@ -72,8 +72,6 @@ task:
kwargs:
ana_long_short: True
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

28
examples/benchmarks/Localformer/workflow_config_localformer_Alpha158.yaml
View File

@@ -34,19 +34,23 @@ data_handler_config: &data_handler_config
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy.strategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
verbose: False
limit_threshold: 0.095
start_time: 2017-01-01
end_time: 2020-08-01
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
exchange_kwargs:
limit_threshold: 0.095
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: LocalformerModel
@@ -70,13 +74,15 @@ task:
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs: {}
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config
config: *port_analysis_config

44
examples/benchmarks/Localformer/workflow_config_localformer_Alpha360.yaml
View File

@@ -26,19 +26,23 @@ data_handler_config: &data_handler_config
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy.strategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
verbose: False
limit_threshold: 0.095
start_time: 2017-01-01
end_time: 2020-08-01
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
exchange_kwargs:
limit_threshold: 0.095
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: LocalformerModel
@@ -59,15 +63,17 @@ task:
valid: [2015-01-01, 2016-12-31]
test: [2017-01-01, 2020-08-01]
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs: {}
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config

2
examples/benchmarks/MLP/workflow_config_mlp_Alpha158.yaml
View File

@@ -95,8 +95,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/MLP/workflow_config_mlp_Alpha360.yaml
View File

@@ -82,8 +82,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

3
examples/benchmarks/README.md
View File

@@ -25,6 +25,7 @@ The numbers shown below demonstrate the performance of the entire `workflow` of
| TCTS (Xueqing Wu, et al.)| Alpha360 | 0.0485±0.00 | 0.3689±0.04| 0.0586±0.00 | 0.4669±0.02 | 0.0816±0.02 | 1.1572±0.30| -0.0689±0.02 |
| Transformer (Ashish Vaswani, et al.)| Alpha360 | 0.0141±0.00 | 0.0917±0.02| 0.0331±0.00 | 0.2357±0.03 | -0.0259±0.03 | -0.3323±0.43| -0.1763±0.07 |
| Localformer (Juyong Jiang, et al.)| Alpha360 | 0.0408±0.00 | 0.2988±0.03| 0.0538±0.00 | 0.4105±0.02 | 0.0275±0.03 | 0.3464±0.37| -0.1182±0.03 |
| TRA (Hengxu Lin, et al.)| Alpha360 | 0.0491±0.01 | 0.3868±0.06 | 0.0589±0.00 | 0.4802±0.04 | 0.0898±0.02 | 1.2490±0.32 | -0.0778±0.02 |
## Alpha158 dataset
| Model Name | Dataset | IC | ICIR | Rank IC | Rank ICIR | Annualized Return | Information Ratio | Max Drawdown |
@@ -43,6 +44,8 @@ The numbers shown below demonstrate the performance of the entire `workflow` of
| TabNet (Sercan O. Arik, et al.)| Alpha158 | 0.0383±0.00 | 0.3414±0.00| 0.0388±0.00 | 0.3460±0.00 | 0.0226±0.00 | 0.2652±0.00| -0.1072±0.00 |
| Transformer (Ashish Vaswani, et al.)| Alpha158 | 0.0274±0.00 | 0.2166±0.04| 0.0409±0.00 | 0.3342±0.04 | 0.0204±0.03 | 0.2888±0.40| -0.1216±0.04 |
| Localformer (Juyong Jiang, et al.)| Alpha158 | 0.0355±0.00 | 0.2747±0.04| 0.0466±0.00 | 0.3762±0.03 | 0.0506±0.02 | 0.7447±0.34| -0.0875±0.02 |
| TRA (Hengxu Lin, et al.)| Alpha158 (with selected 20 features)| 0.0409±0.00 | 0.3253±0.04 | 0.0488±0.00 | 0.4045±0.02 | 0.0673±0.02 | 1.0389±0.39 | -0.0830±0.02 |
| TRA (Hengxu Lin, et al.)| Alpha158 | 0.0442±0.00 | 0.3426±0.03 | 0.0555±0.00 | 0.4395±0.03 | 0.0833±0.03 | 1.2064±0.36 | -0.0849±0.02 |
- The selected 20 features are based on the feature importance of a lightgbm-based model.
- The base model of DoubleEnsemble is LGBM.

2
examples/benchmarks/SFM/requirements.txt
View File

@@ -1,4 +1,4 @@
pandas==1.1.2
numpy==1.17.4
scikit_learn==0.23.2
torch==1.7.0
torch==1.7.0

2
examples/benchmarks/SFM/workflow_config_sfm_Alpha360.yaml
View File

@@ -85,8 +85,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

4
examples/benchmarks/TCTS/requirements.txt Normal file
View File

@@ -0,0 +1,4 @@
pandas==1.1.2
numpy==1.17.4
scikit_learn==0.23.2
torch==1.7.0

2
examples/benchmarks/TCTS/workflow_config_tcts_Alpha360.yaml
View File

@@ -90,8 +90,6 @@ task:
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
ana_long_short: False
ann_scaler: 252
label_col: 1

2
examples/benchmarks/TFT/README.md
View File

@@ -8,7 +8,7 @@
Users can follow the ``workflow_by_code_tft.py`` to run the benchmark.
### Notes
1. Please be **aware** that this script can only support `Python 3.5 - 3.8`.
1. Please be **aware** that this script can only support `Python 3.6 - 3.7`.
2. If the CUDA version on your machine is not 10.0, please remember to run the following commands `conda install anaconda cudatoolkit=10.0` and `conda install cudnn` on your machine.
3. The model must run in GPU, or an error will be raised.
4. New datasets should be registered in ``data_formatters``, for detail please visit the source.

3
examples/benchmarks/TFT/requirements.txt
View File

@@ -1,3 +1,2 @@
tensorflow-gpu==1.15.0
numpy == 1.19.4
pandas==1.1.0
pandas==1.1.0

26
examples/benchmarks/TFT/tft.py
View File

@@ -1,6 +1,8 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
from pathlib import Path
from typing import Union
import numpy as np
import pandas as pd
import tensorflow.compat.v1 as tf
@@ -243,7 +245,7 @@ class TFTModel(ModelFT):
# extract_numerical_data(targets), extract_numerical_data(p90_forecast),
# 0.9)
tf.keras.backend.set_session(default_keras_session)
print("Training completed.".format(dte.datetime.now()))
print("Training completed at {}.".format(dte.datetime.now()))
# ===========================Training Process===========================
def predict(self, dataset):
@@ -289,3 +291,25 @@ class TFTModel(ModelFT):
dataset for finetuning
"""
pass
def to_pickle(self, path: Union[Path, str]):
"""
Tensorflow model can't be dumped directly.
So the data should be save seperatedly
**TODO**: Please implement the function to load the files
Parameters
----------
path : Union[Path, str]
the target path to be dumped
"""
# FIXME: implementing saving tensorflow models
# save tensorflow model
# path = Path(path)
# path.mkdir(parents=True)
# self.model.save(path)
# save qlib model wrapper
self.model = None
super(TFTModel, self).to_pickle(path)

2
examples/benchmarks/TFT/workflow_config_tft_Alpha158.yaml
View File

@@ -58,8 +58,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

94
examples/benchmarks/TRA/README.md
View File

@@ -1,53 +1,78 @@
# Learning Multiple Stock Trading Patterns with Temporal Routing Adaptor and Optimal Transport
This code provides a PyTorch implementation for TRA (Temporal Routing Adaptor), as described in the paper [Learning Multiple Stock Trading Patterns with Temporal Routing Adaptor and Optimal Transport](http://arxiv.org/abs/2106.12950).
Temporal Routing Adaptor (TRA) is designed to capture multiple trading patterns in the stock market data. Please refer to [our paper](http://arxiv.org/abs/2106.12950) for more details.
* TRA (Temporal Routing Adaptor) is a lightweight module that consists of a set of independent predictors for learning multiple patterns as well as a router to dispatch samples to different predictors.
* We also design a learning algorithm based on Optimal Transport (OT) to obtain the optimal sample to predictor assignment and effectively optimize the router with such assignment through an auxiliary loss term.
If you find our work useful in your research, please cite:
```
@inproceedings{HengxuKDD2021,
author = {Hengxu Lin and Dong Zhou and Weiqing Liu and Jiang Bian},
title = {Learning Multiple Stock Trading Patterns with Temporal Routing Adaptor and Optimal Transport},
booktitle = {Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery \& Data Mining},
series = {KDD '21},
year = {2021},
publisher = {ACM},
}
@article{yang2020qlib,
title={Qlib: An AI-oriented Quantitative Investment Platform},
author={Yang, Xiao and Liu, Weiqing and Zhou, Dong and Bian, Jiang and Liu, Tie-Yan},
journal={arXiv preprint arXiv:2009.11189},
year={2020}
}
```
# Running TRA
## Usage (Recommended)
## Requirements
- Install `Qlib` main branch
**Update**: `TRA` has been moved to `qlib.contrib.model.pytorch_tra` to support other `Qlib` components like `qlib.workflow` and `Alpha158/Alpha360` dataset.
## Running
Please follow the official [doc](https://qlib.readthedocs.io/en/latest/component/workflow.html) to use `TRA` with `workflow`. Here we also provide several example config files:
- `workflow_config_tra_Alpha360.yaml`: running `TRA` with `Alpha360` dataset
- `workflow_config_tra_Alpha158.yaml`: running `TRA` with `Alpha158` dataset (with feature subsampling)
- `workflow_config_tra_Alpha158_full.yaml`: running `TRA` with `Alpha158` dataset (without feature subsampling)
The performances of `TRA` are reported in [Benchmarks](https://github.com/microsoft/qlib/tree/main/examples/benchmarks).
## Usage (Not Maintained)
This section is used to reproduce the results in the paper.
### Running
We attach our running scripts for the paper in `run.sh`.
And here are two ways to run the model:
* Running from scripts with default parameters
You can directly run from Qlib command `qrun`:
```
qrun configs/config_alstm.yaml
```
You can directly run from Qlib command `qrun`:
```
qrun configs/config_alstm.yaml
```
* Running from code with self-defined parameters
Setting different parameters is also allowed. See codes in `example.py`:
```
python example.py --config_file configs/config_alstm.yaml
```
Setting different parameters is also allowed. See codes in `example.py`:
```
python example.py --config_file configs/config_alstm.yaml
```
Here we trained TRA on a pretrained backbone model. Therefore we run `*_init.yaml` before TRA's scipts.
# Results
## Outputs
### Results
After running the scripts, you can find result files in path `./output`:
`info.json` - config settings and result metrics.
* `info.json` - config settings and result metrics.
* `log.csv` - running logs.
* `model.bin` - the model parameter dictionary.
* `pred.pkl` - the prediction scores and output for inference.
`log.csv` - running logs.
Evaluation metrics reported in the paper:
This result is generated by qlib==0.7.1.
`model.bin` - the model parameter dictionary.
`pred.pkl` - the prediction scores and output for inference.
## Our Results
| Methods | MSE| MAE| IC | ICIR | AR | AV | SR | MDD |
|-------------------|-------------------|---------------------|--------------------|--------------------|--------------------|--------------------|--------------------|--------------------|
|-------|-------|------|-----|-----|-----|-----|-----|-----|
|Linear|0.163|0.327|0.020|0.132|-3.2%|16.8%|-0.191|32.1%|
|LightGBM|0.160(0.000)|0.323(0.000)|0.041|0.292|7.8%|15.5%|0.503|25.7%|
|MLP|0.160(0.002)|0.323(0.003)|0.037|0.273|3.7%|15.3%|0.264|26.2%|
@@ -61,21 +86,8 @@ After running the scripts, you can find result files in path `./output`:
A more detailed demo for our experiment results in the paper can be found in `Report.ipynb`.
# Common Issues
## Common Issues
For help or issues using TRA, please submit a GitHub issue.
Sometimes we might encounter situation where the loss is `NaN`, please check the `epsilon` parameter in the sinkhorn algorithm, adjusting the `epsilon` according to input's scale is important.
# Citation
If you find this repository useful in your research, please cite:
```
@inproceedings{HengxuKDD2021,
author = {Hengxu Lin and Dong Zhou and Weiqing Liu and Jiang Bian},
title = {Learning Multiple Stock Trading Patterns with Temporal Routing Adaptor and Optimal Transport},
booktitle = {Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery \& Data Mining},
series = {KDD '21},
year = {2021},
publisher = {ACM},
}
```
Sometimes we might encounter situation where the loss is `NaN`, please check the `epsilon` parameter in the sinkhorn algorithm, adjusting the `epsilon` according to input's scale is important.

5
examples/benchmarks/TRA/requirements.txt Normal file
View File

@@ -0,0 +1,5 @@
pandas==1.1.2
numpy==1.17.4
scikit_learn==0.23.2
torch==1.7.0
seaborn

132
examples/benchmarks/TRA/workflow_config_tra_Alpha158.yaml Normal file
View File

@@ -0,0 +1,132 @@
qlib_init:
provider_uri: "~/.qlib/qlib_data/cn_data"
region: cn
market: &market csi300
benchmark: &benchmark SH000300
data_handler_config: &data_handler_config
start_time: 2008-01-01
end_time: 2020-08-01
fit_start_time: 2008-01-01
fit_end_time: 2014-12-31
instruments: *market
infer_processors:
- class: FilterCol
kwargs:
fields_group: feature
col_list: ["RESI5", "WVMA5", "RSQR5", "KLEN", "RSQR10", "CORR5", "CORD5", "CORR10",
"ROC60", "RESI10", "VSTD5", "RSQR60", "CORR60", "WVMA60", "STD5",
"RSQR20", "CORD60", "CORD10", "CORR20", "KLOW"]
- class: RobustZScoreNorm
kwargs:
fields_group: feature
clip_outlier: true
- class: Fillna
kwargs:
fields_group: feature
learn_processors:
- class: CSRankNorm
kwargs:
fields_group: label
label: ["Ref($close, -2) / Ref($close, -1) - 1"]
num_states: &num_states 3
memory_mode: &memory_mode sample
tra_config: &tra_config
num_states: *num_states
rnn_arch: LSTM
hidden_size: 32
num_layers: 1
dropout: 0.0
tau: 1.0
src_info: LR_TPE
model_config: &model_config
input_size: 20
hidden_size: 64
num_layers: 2
rnn_arch: LSTM
use_attn: True
dropout: 0.0
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
limit_threshold: 0.095
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: TRAModel
module_path: qlib.contrib.model.pytorch_tra
kwargs:
tra_config: *tra_config
model_config: *model_config
model_type: RNN
lr: 1e-3
n_epochs: 100
max_steps_per_epoch:
early_stop: 20
logdir: output/Alpha158
seed: 0
lamb: 1.0
rho: 0.99
alpha: 0.5
transport_method: router
memory_mode: *memory_mode
eval_train: False
eval_test: True
pretrain: True
init_state:
freeze_model: False
freeze_predictors: False
dataset:
class: MTSDatasetH
module_path: qlib.contrib.data.dataset
kwargs:
handler:
class: Alpha158
module_path: qlib.contrib.data.handler
kwargs: *data_handler_config
segments:
train: [2008-01-01, 2014-12-31]
valid: [2015-01-01, 2016-12-31]
test: [2017-01-01, 2020-08-01]
seq_len: 60
horizon: 2
input_size:
num_states: *num_states
batch_size: 1024
n_samples:
memory_mode: *memory_mode
drop_last: True
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config

126
examples/benchmarks/TRA/workflow_config_tra_Alpha158_full.yaml Normal file
View File

@@ -0,0 +1,126 @@
qlib_init:
provider_uri: "~/.qlib/qlib_data/cn_data"
region: cn
market: &market csi300
benchmark: &benchmark SH000300
data_handler_config: &data_handler_config
start_time: 2008-01-01
end_time: 2020-08-01
fit_start_time: 2008-01-01
fit_end_time: 2014-12-31
instruments: *market
infer_processors:
- class: RobustZScoreNorm
kwargs:
fields_group: feature
clip_outlier: true
- class: Fillna
kwargs:
fields_group: feature
learn_processors:
- class: CSRankNorm
kwargs:
fields_group: label
label: ["Ref($close, -2) / Ref($close, -1) - 1"]
num_states: &num_states 3
memory_mode: &memory_mode sample
tra_config: &tra_config
num_states: *num_states
rnn_arch: LSTM
hidden_size: 32
num_layers: 1
dropout: 0.0
tau: 1.0
src_info: LR_TPE
model_config: &model_config
input_size: 158
hidden_size: 256
num_layers: 2
rnn_arch: LSTM
use_attn: True
dropout: 0.2
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
limit_threshold: 0.095
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: TRAModel
module_path: qlib.contrib.model.pytorch_tra
kwargs:
tra_config: *tra_config
model_config: *model_config
model_type: RNN
lr: 1e-3
n_epochs: 100
max_steps_per_epoch:
early_stop: 20
logdir: output/Alpha158_full
seed: 0
lamb: 1.0
rho: 0.99
alpha: 0.5
transport_method: router
memory_mode: *memory_mode
eval_train: False
eval_test: True
pretrain: True
init_state:
freeze_model: False
freeze_predictors: False
dataset:
class: MTSDatasetH
module_path: qlib.contrib.data.dataset
kwargs:
handler:
class: Alpha158
module_path: qlib.contrib.data.handler
kwargs: *data_handler_config
segments:
train: [2008-01-01, 2014-12-31]
valid: [2015-01-01, 2016-12-31]
test: [2017-01-01, 2020-08-01]
seq_len: 60
horizon: 2
input_size:
num_states: *num_states
batch_size: 1024
n_samples:
memory_mode: *memory_mode
drop_last: True
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config

126
examples/benchmarks/TRA/workflow_config_tra_Alpha360.yaml Normal file
View File

@@ -0,0 +1,126 @@
qlib_init:
provider_uri: "~/.qlib/qlib_data/cn_data"
region: cn
market: &market csi300
benchmark: &benchmark SH000300
data_handler_config: &data_handler_config
start_time: 2008-01-01
end_time: 2020-08-01
fit_start_time: 2008-01-01
fit_end_time: 2014-12-31
instruments: *market
infer_processors:
- class: RobustZScoreNorm
kwargs:
fields_group: feature
clip_outlier: true
- class: Fillna
kwargs:
fields_group: feature
learn_processors:
- class: CSRankNorm
kwargs:
fields_group: label
label: ["Ref($close, -2) / Ref($close, -1) - 1"]
num_states: &num_states 3
memory_mode: &memory_mode sample
tra_config: &tra_config
num_states: *num_states
rnn_arch: LSTM
hidden_size: 32
num_layers: 1
dropout: 0.0
tau: 1.0
src_info: LR_TPE
model_config: &model_config
input_size: 6
hidden_size: 64
num_layers: 2
rnn_arch: LSTM
use_attn: True
dropout: 0.0
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
limit_threshold: 0.095
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: TRAModel
module_path: qlib.contrib.model.pytorch_tra
kwargs:
tra_config: *tra_config
model_config: *model_config
model_type: RNN
lr: 1e-3
n_epochs: 100
max_steps_per_epoch:
early_stop: 20
logdir: output/Alpha360
seed: 0
lamb: 1.0
rho: 0.99
alpha: 0.5
transport_method: router
memory_mode: *memory_mode
eval_train: False
eval_test: True
pretrain: True
init_state:
freeze_model: False
freeze_predictors: False
dataset:
class: MTSDatasetH
module_path: qlib.contrib.data.dataset
kwargs:
handler:
class: Alpha360
module_path: qlib.contrib.data.handler
kwargs: *data_handler_config
segments:
train: [2008-01-01, 2014-12-31]
valid: [2015-01-01, 2016-12-31]
test: [2017-01-01, 2020-08-01]
seq_len: 60
horizon: 2
input_size: 6
num_states: *num_states
batch_size: 1024
n_samples:
memory_mode: *memory_mode
drop_last: True
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config

2
examples/benchmarks/TabNet/workflow_config_TabNet_Alpha158.yaml
View File

@@ -75,8 +75,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/TabNet/workflow_config_TabNet_Alpha360.yaml
View File

@@ -75,8 +75,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

22
examples/benchmarks/Transformer/workflow_config_transformer_Alpha158.yaml
View File

@@ -34,19 +34,23 @@ data_handler_config: &data_handler_config
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy.strategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
verbose: False
limit_threshold: 0.095
start_time: 2017-01-01
end_time: 2020-08-01
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
exchange_kwargs:
limit_threshold: 0.095
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: TransformerModel
@@ -70,7 +74,9 @@ task:
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs: {}
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

24
examples/benchmarks/Transformer/workflow_config_transformer_Alpha360.yaml
View File

@@ -26,19 +26,23 @@ data_handler_config: &data_handler_config
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy.strategy
module_path: qlib.contrib.strategy
kwargs:
model: <MODEL>
dataset: <DATASET>
topk: 50
n_drop: 5
backtest:
verbose: False
limit_threshold: 0.095
start_time: 2017-01-01
end_time: 2020-08-01
account: 100000000
benchmark: *benchmark
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
exchange_kwargs:
limit_threshold: 0.095
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: TransformerModel
@@ -61,7 +65,9 @@ task:
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs: {}
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
@@ -70,4 +76,4 @@ task:
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config
config: *port_analysis_config

2
examples/benchmarks/XGBoost/workflow_config_xgboost_Alpha158.yaml
View File

@@ -64,8 +64,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/benchmarks/XGBoost/workflow_config_xgboost_Alpha360.yaml
View File

@@ -71,8 +71,6 @@ task:
kwargs:
ana_long_short: False
ann_scaler: 252
model: <MODEL>
dataset: <DATASET>
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:

2
examples/highfreq/workflow_config_High_Freq_Tree_Alpha158.yaml
View File

@@ -60,8 +60,6 @@ task:
- class: "SignalRecord"
module_path: "qlib.workflow.record_temp"
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: "HFSignalRecord"
module_path: "qlib.workflow.record_temp"
kwargs: {}

1
examples/model_rolling/requirements.txt Normal file
View File

@@ -0,0 +1 @@
xgboost

2
examples/model_rolling/task_manager_rolling.py
View File

@@ -17,7 +17,7 @@ from qlib.workflow.task.gen import RollingGen, task_generator
from qlib.workflow.task.manage import TaskManager, run_task
from qlib.workflow.task.collect import RecorderCollector
from qlib.model.ens.group import RollingGroup
from qlib.model.trainer import TrainerRM
from qlib.model.trainer import TrainerRM, task_train
from qlib.tests.config import CSI100_RECORD_LGB_TASK_CONFIG, CSI100_RECORD_XGBOOST_TASK_CONFIG

50
examples/nested_decision_execution/workflow.py
View File

@@ -19,7 +19,7 @@ class NestedDecisionExecutionWorkflow:
benchmark = "SH000300"
data_handler_config = {
"start_time": "2008-01-01",
"end_time": "2020-12-31",
"end_time": "2021-05-31",
"fit_start_time": "2008-01-01",
"fit_end_time": "2014-12-31",
"instruments": market,
@@ -53,7 +53,7 @@ class NestedDecisionExecutionWorkflow:
"segments": {
"train": ("2007-01-01", "2014-12-31"),
"valid": ("2015-01-01", "2016-12-31"),
"test": ("2020-01-01", "2020-12-31"),
"test": ("2020-01-01", "2021-05-31"),
},
},
},
@@ -75,7 +75,7 @@ class NestedDecisionExecutionWorkflow:
"module_path": "qlib.backtest.executor",
"kwargs": {
"time_per_step": "5min",
"generate_report": True,
"generate_portfolio_metrics": True,
"verbose": True,
"indicator_config": {
"show_indicator": True,
@@ -86,7 +86,7 @@ class NestedDecisionExecutionWorkflow:
"class": "TWAPStrategy",
"module_path": "qlib.contrib.strategy.rule_strategy",
},
"generate_report": True,
"generate_portfolio_metrics": True,
"indicator_config": {
"show_indicator": True,
},
@@ -101,15 +101,15 @@ class NestedDecisionExecutionWorkflow:
},
},
"track_data": True,
"generate_report": True,
"generate_portfolio_metrics": True,
"indicator_config": {
"show_indicator": True,
},
},
},
"backtest": {
"start_time": "2020-01-01",
"end_time": "2020-12-31",
"start_time": "2020-09-20",
"end_time": "2021-05-20",
"account": 100000000,
"exchange_kwargs": {
"freq": "1min",
@@ -124,8 +124,6 @@ class NestedDecisionExecutionWorkflow:
def _init_qlib(self):
"""initialize qlib"""
# provider_uri_day = "/data/stock_data/huaxia/qlib"
# provider_uri_1min = "/data2/stock_data/huaxia_1min_qlib"
provider_uri_day = "~/.qlib/qlib_data/cn_data" # target_dir
GetData().qlib_data(target_dir=provider_uri_day, region=REG_CN, version="v2", exists_skip=True)
provider_uri_1min = HIGH_FREQ_CONFIG.get("provider_uri")
@@ -133,31 +131,7 @@ class NestedDecisionExecutionWorkflow:
target_dir=provider_uri_1min, interval="1min", region=REG_CN, version="v2", exists_skip=True
)
provider_uri_map = {"1min": provider_uri_1min, "day": provider_uri_day}
client_config = {
"calendar_provider": {
"class": "LocalCalendarProvider",
"module_path": "qlib.data.data",
"kwargs": {
"backend": {
"class": "FileCalendarStorage",
"module_path": "qlib.data.storage.file_storage",
"kwargs": {"provider_uri_map": provider_uri_map},
}
},
},
"feature_provider": {
"class": "LocalFeatureProvider",
"module_path": "qlib.data.data",
"kwargs": {
"backend": {
"class": "FileFeatureStorage",
"module_path": "qlib.data.storage.file_storage",
"kwargs": {"provider_uri_map": provider_uri_map},
}
},
},
}
qlib.init(provider_uri=provider_uri_day, **client_config, redis_port=-1)
qlib.init(provider_uri=provider_uri_map, dataset_cache=None, expression_cache=None)
def _train_model(self, model, dataset):
with R.start(experiment_name="train"):
@@ -186,9 +160,8 @@ class NestedDecisionExecutionWorkflow:
},
}
self.port_analysis_config["strategy"] = strategy_config
self.port_analysis_config["backtest"]["benchmark"] = D.list_instruments(
instruments=D.instruments(market=self.market), as_list=True
)
self.port_analysis_config["backtest"]["benchmark"] = self.benchmark
with R.start(experiment_name="backtest"):
recorder = R.get_recorder()
@@ -201,6 +174,7 @@ class NestedDecisionExecutionWorkflow:
)
par.generate()
# user could use following methods to analysis the position
# report_normal_df = recorder.load_object("portfolio_analysis/report_normal_1day.pkl")
# from qlib.contrib.report import analysis_position
# analysis_position.report_graph(report_normal_df)
@@ -212,7 +186,7 @@ class NestedDecisionExecutionWorkflow:
self._train_model(model, dataset)
executor_config = self.port_analysis_config["executor"]
backtest_config = self.port_analysis_config["backtest"]
backtest_config["benchmark"] = D.list_instruments(instruments=D.instruments(market=self.market), as_list=True)
backtest_config["benchmark"] = self.benchmark
strategy_config = {
"class": "TopkDropoutStrategy",
"module_path": "qlib.contrib.strategy.model_strategy",

125
examples/run_all_model.py
View File

@@ -6,6 +6,7 @@ import sys
import fire
import time
import glob
import yaml
import shutil
import signal
import inspect
@@ -23,22 +24,6 @@ from qlib.config import REG_CN
from qlib.workflow import R
from qlib.tests.data import GetData
# init qlib
provider_uri = "~/.qlib/qlib_data/cn_data"
exp_folder_name = "run_all_model_records"
exp_path = str(Path(os.getcwd()).resolve() / exp_folder_name)
exp_manager = {
"class": "MLflowExpManager",
"module_path": "qlib.workflow.expm",
"kwargs": {
"uri": "file:" + exp_path,
"default_exp_name": "Experiment",
},
}
GetData().qlib_data(target_dir=provider_uri, region=REG_CN, exists_skip=True)
qlib.init(provider_uri=provider_uri, region=REG_CN, exp_manager=exp_manager)
# decorator to check the arguments
def only_allow_defined_args(function_to_decorate):
@@ -88,11 +73,11 @@ def create_env():
sys.stderr.write("\n")
# get anaconda activate path
conda_activate = Path(os.environ["CONDA_PREFIX"]) / "bin" / "activate" # TODO: FIX ME!
return env_path, python_path, conda_activate
return temp_dir, env_path, python_path, conda_activate
# function to execute the cmd
def execute(cmd, wait_when_err=False):
def execute(cmd, wait_when_err=False, raise_err=True):
print("Running CMD:", cmd)
with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True, shell=True) as p:
for line in p.stdout:
@@ -105,6 +90,8 @@ def execute(cmd, wait_when_err=False):
if p.returncode != 0:
if wait_when_err:
input("Press Enter to Continue")
if raise_err:
raise RuntimeError(f"Error when executing command: {cmd}")
return p.stderr
else:
return None
@@ -134,14 +121,23 @@ def get_all_folders(models, exclude) -> dict:
def get_all_files(folder_path, dataset) -> (str, str):
yaml_path = str(Path(f"{folder_path}") / f"*{dataset}*.yaml")
req_path = str(Path(f"{folder_path}") / f"*.txt")
return glob.glob(yaml_path)[0], glob.glob(req_path)[0]
yaml_file = glob.glob(yaml_path)
req_file = glob.glob(req_path)
if len(yaml_file) == 0:
return None, None
else:
return yaml_file[0], req_file[0]
# function to retrieve all the results
def get_all_results(folders) -> dict:
results = dict()
for fn in folders:
exp = R.get_exp(experiment_name=fn, create=False)
try:
exp = R.get_exp(experiment_name=fn, create=False)
except ValueError:
# No experiment results
continue
recorders = exp.list_recorders()
result = dict()
result["annualized_return_with_cost"] = list()
@@ -155,9 +151,9 @@ def get_all_results(folders) -> dict:
if recorders[recorder_id].status == "FINISHED":
recorder = R.get_recorder(recorder_id=recorder_id, experiment_name=fn)
metrics = recorder.list_metrics()
result["annualized_return_with_cost"].append(metrics["excess_return_with_cost.annualized_return"])
result["information_ratio_with_cost"].append(metrics["excess_return_with_cost.information_ratio"])
result["max_drawdown_with_cost"].append(metrics["excess_return_with_cost.max_drawdown"])
result["annualized_return_with_cost"].append(metrics["1day.excess_return_with_cost.annualized_return"])
result["information_ratio_with_cost"].append(metrics["1day.excess_return_with_cost.information_ratio"])
result["max_drawdown_with_cost"].append(metrics["1day.excess_return_with_cost.max_drawdown"])
result["ic"].append(metrics["IC"])
result["icir"].append(metrics["ICIR"])
result["rank_ic"].append(metrics["Rank IC"])
@@ -185,6 +181,25 @@ def gen_and_save_md_table(metrics, dataset):
return table
# read yaml, remove seed kwargs of model, and then save file in the temp_dir
def gen_yaml_file_without_seed_kwargs(yaml_path, temp_dir):
with open(yaml_path, "r") as fp:
config = yaml.load(fp)
try:
del config["task"]["model"]["kwargs"]["seed"]
except KeyError:
# If the key does not exists, use original yaml
# NOTE: it is very important if the model most run in original path(when sys.rel_path is used)
return yaml_path
else:
# otherwise, generating a new yaml without random seed
file_name = yaml_path.split("/")[-1]
temp_path = os.path.join(temp_dir, file_name)
with open(temp_path, "w") as fp:
yaml.dump(config, fp)
return temp_path
# function to run the all the models
@only_allow_defined_args
def run(
@@ -193,12 +208,13 @@ def run(
dataset="Alpha360",
exclude=False,
qlib_uri: str = "git+https://github.com/microsoft/qlib#egg=pyqlib",
exp_folder_name: str = "run_all_model_records",
wait_before_rm_env: bool = False,
wait_when_err: bool = False,
):
"""
Please be aware that this function can only work under Linux. MacOS and Windows will be supported in the future.
Any PR to enhance this method is highly welcomed. Besides, this script doesn't support parrallel running the same model
Any PR to enhance this method is highly welcomed. Besides, this script doesn't support parallel running the same model
for multiple times, and this will be fixed in the future development.
Parameters:
@@ -214,6 +230,8 @@ def run(
qlib_uri : str
the uri to install qlib with pip
it could be url on the we or local path
exp_folder_name: str
the name of the experiment folder
wait_before_rm_env : bool
wait before remove environment.
wait_when_err : bool
@@ -240,26 +258,58 @@ def run(
# Case 5 - run specific models for one time
python run_all_model.py --models=[mlp,lightgbm]
# Case 6 - run other models except those are given as aruments for one time
# Case 6 - run other models except those are given as arguments for one time
python run_all_model.py --models=[mlp,tft,sfm] --exclude=True
"""
# init qlib
GetData().qlib_data(exists_skip=True)
qlib.init(
exp_manager={
"class": "MLflowExpManager",
"module_path": "qlib.workflow.expm",
"kwargs": {
"uri": "file:" + str(Path(os.getcwd()).resolve() / exp_folder_name),
"default_exp_name": "Experiment",
},
}
)
# get all folders
folders = get_all_folders(models, exclude)
# init error messages:
errors = dict()
# run all the model for iterations
for fn in folders:
# create env by anaconda
env_path, python_path, conda_activate = create_env()
# get all files
sys.stderr.write("Retrieving files...\n")
yaml_path, req_path = get_all_files(folders[fn], dataset)
if yaml_path is None:
sys.stderr.write(f"There is no {dataset}.yaml file in {folders[fn]}")
continue
sys.stderr.write("\n")
# create env by anaconda
temp_dir, env_path, python_path, conda_activate = create_env()
# install requirements.txt
sys.stderr.write("Installing requirements.txt...\n")
execute(f"{python_path} -m pip install -r {req_path}", wait_when_err=wait_when_err)
with open(req_path) as f:
content = f.read()
if "torch" in content:
# automatically install pytorch according to nvidia's version
execute(
f"{python_path} -m pip install light-the-torch", wait_when_err=wait_when_err
) # for automatically installing torch according to the nvidia driver
execute(
f"{env_path / 'bin' / 'ltt'} install --install-cmd '{python_path} -m pip install {{packages}}' -- -r {req_path}",
wait_when_err=wait_when_err,
)
else:
execute(f"{python_path} -m pip install -r {req_path}", wait_when_err=wait_when_err)
sys.stderr.write("\n")
# read yaml, remove seed kwargs of model, and then save file in the temp_dir
yaml_path = gen_yaml_file_without_seed_kwargs(yaml_path, temp_dir)
# setup gpu for tft
if fn == "TFT":
execute(
@@ -302,19 +352,20 @@ def run(
# getting all results
sys.stderr.write(f"Retrieving results...\n")
results = get_all_results(folders)
# calculating the mean and std
sys.stderr.write(f"Calculating the mean and std of results...\n")
results = cal_mean_std(results)
# generating md table
sys.stderr.write(f"Generating markdown table...\n")
gen_and_save_md_table(results, dataset)
sys.stderr.write("\n")
# print erros
if len(results) > 0:
# calculating the mean and std
sys.stderr.write(f"Calculating the mean and std of results...\n")
results = cal_mean_std(results)
# generating md table
sys.stderr.write(f"Generating markdown table...\n")
gen_and_save_md_table(results, dataset)
sys.stderr.write("\n")
# print errors
sys.stderr.write(f"Here are some of the errors of the models...\n")
pprint(errors)
sys.stderr.write("\n")
# move results folder
shutil.move(exp_path, exp_path + f"_{dataset}_{datetime.now().strftime('%Y-%m-%d_%H:%M:%S')}")
shutil.move(exp_folder_name, exp_folder_name + f"_{dataset}_{datetime.now().strftime('%Y-%m-%d_%H:%M:%S')}")
shutil.move("table.md", f"table_{dataset}_{datetime.now().strftime('%Y-%m-%d_%H:%M:%S')}.md")

13
examples/workflow_by_code.ipynb
View File

@@ -20,9 +20,7 @@
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"scrolled": true
},
"metadata": {},
"outputs": [],
"source": [
"import sys, site\n",
@@ -201,7 +199,7 @@
" \"module_path\": \"qlib.backtest.executor\",\n",
" \"kwargs\": {\n",
" \"time_per_step\": \"day\",\n",
" \"generate_report\": True,\n",
" \"generate_portfolio_metrics\": True,\n",
" },\n",
" },\n",
" \"strategy\": {\n",
@@ -362,7 +360,7 @@
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
@@ -375,8 +373,7 @@
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.3"
"pygments_lexer": "ipython3"
},
"toc": {
"base_numbering": 1,
@@ -394,4 +391,4 @@
},
"nbformat": 4,
"nbformat_minor": 4
}
}

2
examples/workflow_by_code.py
View File

@@ -26,7 +26,7 @@ if __name__ == "__main__":
"module_path": "qlib.backtest.executor",
"kwargs": {
"time_per_step": "day",
"generate_report": True,
"generate_portfolio_metrics": True,
},
},
"strategy": {