def end_task_train(rec: Recorder, experiment_name: str) -> Recorder:
"""
Finish task training with real model fitting and saving.
Args:
rec (Recorder): the recorder will be resumed
experiment_name (str): the name of experiment
Returns:
Recorder: the model recorder
"""
with R.start(experiment_name=experiment_name,
recorder_id=rec.info["id"],
resume=True):
task_config = R.load_object("task")
# model & dataset initiation
model: Model = init_instance_by_config(task_config["model"])
dataset: Dataset = init_instance_by_config(task_config["dataset"])
# model training
model.fit(dataset)
R.save_objects(**{"params.pkl": model})
# this dataset is saved for online inference. So the concrete data should not be dumped
dataset.config(dump_all=False, recursive=True)
R.save_objects(**{"dataset": dataset})
# fill placehorder
placehorder_value = {"<MODEL>": model, "<DATASET>": dataset}
task_config = fill_placeholder(task_config, placehorder_value)
# generate records: prediction, backtest, and analysis
records = task_config.get("record", [])
if isinstance(records, dict): # prevent only one dict
records = [records]
for record in records:
r = init_instance_by_config(record, recorder=rec)
r.generate()
return rec
def _exe_task(task_config: dict):
rec = R.get_recorder()
# model & dataset initiation
model: Model = init_instance_by_config(task_config["model"])
dataset: Dataset = init_instance_by_config(task_config["dataset"])
# FIXME: resume reweighter after merging data selection
# reweighter: Reweighter = task_config.get("reweighter", None)
# model training
# auto_filter_kwargs(model.fit)(dataset, reweighter=reweighter)
model.fit(dataset)
R.save_objects(**{"params.pkl": model})
# this dataset is saved for online inference. So the concrete data should not be dumped
dataset.config(dump_all=False, recursive=True)
R.save_objects(**{"dataset": dataset})
# fill placehorder
placehorder_value = {"<MODEL>": model, "<DATASET>": dataset}
task_config = fill_placeholder(task_config, placehorder_value)
# generate records: prediction, backtest, and analysis
records = task_config.get("record", [])
if isinstance(records, dict): # prevent only one dict
records = [records]
for record in records:
# Some recorder require the parameter `model` and `dataset`.
# try to automatically pass in them to the initialization function
# to make defining the tasking easier
r = init_instance_by_config(
record,
recorder=rec,
default_module="qlib.workflow.record_temp",
try_kwargs={
"model": model,
"dataset": dataset
},
)
r.generate()
def train_meta_model(self):
"""
training a meta model based on a simplified linear proxy model;
"""
# 1) leverage the simplified proxy forecasting model to train meta model.
# - Only the dataset part is important, in current version of meta model will integrate the
rb = RollingBenchmark(model_type=self.sim_task_model)
sim_task = rb.basic_task()
proxy_forecast_model_task = {
# "model": "qlib.contrib.model.linear.LinearModel",
"dataset": {
"class": "qlib.data.dataset.DatasetH",
"kwargs": {
"handler":
f"file://{(DIRNAME / 'handler_proxy.pkl').absolute()}",
"segments": {
"train": ("2008-01-01", "2010-12-31"),
"test":
("2011-01-01",
sim_task["dataset"]["kwargs"]["segments"]["test"][1]),
},
},
},
# "record": ["qlib.workflow.record_temp.SignalRecord"]
}
# the proxy_forecast_model_task will be used to create meta tasks.
# The test date of first task will be 2011-01-01. Each test segment will be about 20days
# The tasks include all training tasks and test tasks.
# 2) preparing meta dataset
kwargs = dict(
task_tpl=proxy_forecast_model_task,
step=self.step,
segments=0.62, # keep test period consistent with the dataset yaml
trunc_days=1 + self.horizon,
hist_step_n=30,
fill_method="max",
rolling_ext_days=0,
)
# NOTE:
# the input of meta model (internal data) are shared between proxy model and final forecasting model
# but their task test segment are not aligned! It worked in my previous experiment.
# So the misalignment will not affect the effectiveness of the method.
with self._internal_data_path.open("rb") as f:
internal_data = pickle.load(f)
md = MetaDatasetDS(exp_name=internal_data, **kwargs)
# 3) train and logging meta model
with R.start(experiment_name=self.meta_exp_name):
R.log_params(**kwargs)
mm = MetaModelDS(step=self.step,
hist_step_n=kwargs["hist_step_n"],
lr=0.001,
max_epoch=200,
seed=43)
mm.fit(md)
R.save_objects(model=mm)
def _train_model(self, model, dataset):
with R.start(experiment_name="train"):
R.log_params(**flatten_dict(self.task))
model.fit(dataset)
R.save_objects(**{"params.pkl": model})
# prediction
recorder = R.get_recorder()
sr = SignalRecord(model, dataset, recorder)
sr.generate()
def run_exp(task_config, dataset, experiment_name, recorder_name, uri):
model = init_instance_by_config(task_config["model"])
model_fit_kwargs = dict(dataset=dataset)
# Let's start the experiment.
with R.start(
experiment_name=experiment_name,
recorder_name=recorder_name,
uri=uri,
resume=True,
):
# Setup log
recorder_root_dir = R.get_recorder().get_local_dir()
log_file = os.path.join(recorder_root_dir,
"{:}.log".format(experiment_name))
set_log_basic_config(log_file)
logger = get_module_logger("q.run_exp")
logger.info("task_config::\n{:}".format(
pprint.pformat(task_config, indent=2)))
logger.info("[{:}] - [{:}]: {:}".format(experiment_name, recorder_name,
uri))
logger.info("dataset={:}".format(dataset))
# Train model
R.log_params(**flatten_dict(task_config))
if "save_path" in inspect.getfullargspec(model.fit).args:
model_fit_kwargs["save_path"] = os.path.join(
recorder_root_dir, "model.ckp")
elif "save_dir" in inspect.getfullargspec(model.fit).args:
model_fit_kwargs["save_dir"] = os.path.join(
recorder_root_dir, "model-ckps")
model.fit(**model_fit_kwargs)
# Get the recorder
recorder = R.get_recorder()
R.save_objects(**{"model.pkl": model})
# Generate records: prediction, backtest, and analysis
for record in task_config["record"]:
record = record.copy()
if record["class"] == "SignalRecord":
srconf = {
"model": model,
"dataset": dataset,
"recorder": recorder
}
record["kwargs"].update(srconf)
sr = init_instance_by_config(record)
sr.generate()
else:
rconf = {"recorder": recorder}
record["kwargs"].update(rconf)
ar = init_instance_by_config(record)
ar.generate()
def ens_rolling(self):
rc = RecorderCollector(
experiment=self.rolling_exp,
artifacts_key=["pred", "label"],
process_list=[RollingEnsemble()],
# rec_key_func=lambda rec: (self.COMB_EXP, rec.info["id"]),
artifacts_path={
"pred": "pred.pkl",
"label": "label.pkl"
},
)
res = rc()
with R.start(experiment_name=self.COMB_EXP):
R.log_params(exp_name=self.rolling_exp)
R.save_objects(**{
"pred.pkl": res["pred"],
"label.pkl": res["label"]
})
def train(uri_path: str = None):
"""train model
Returns
-------
pred_score: pandas.DataFrame
predict scores
performance: dict
model performance
"""
# model initiaiton
model = init_instance_by_config(CSI300_GBDT_TASK["model"])
dataset = init_instance_by_config(CSI300_GBDT_TASK["dataset"])
# To test __repr__
print(dataset)
print(R)
# start exp
with R.start(experiment_name="workflow", uri=uri_path):
R.log_params(**flatten_dict(CSI300_GBDT_TASK))
model.fit(dataset)
R.save_objects(trained_model=model)
# prediction
recorder = R.get_recorder()
# To test __repr__
print(recorder)
# To test get_local_dir
print(recorder.get_local_dir())
rid = recorder.id
sr = SignalRecord(model, dataset, recorder)
sr.generate()
pred_score = sr.load("pred.pkl")
# calculate ic and ric
sar = SigAnaRecord(recorder)
sar.generate()
ic = sar.load("ic.pkl")
ric = sar.load("ric.pkl")
return pred_score, {"ic": ic, "ric": ric}, rid
def run_exp(task_config, dataset, experiment_name, recorder_name, uri):
# model initiaiton
print("")
print("[{:}] - [{:}]: {:}".format(experiment_name, recorder_name, uri))
print("dataset={:}".format(dataset))
model = init_instance_by_config(task_config["model"])
# start exp
with R.start(experiment_name=experiment_name,
recorder_name=recorder_name,
uri=uri):
log_file = R.get_recorder().root_uri / "{:}.log".format(
experiment_name)
set_log_basic_config(log_file)
# train model
R.log_params(**flatten_dict(task_config))
model.fit(dataset)
recorder = R.get_recorder()
R.save_objects(**{"model.pkl": model})
# generate records: prediction, backtest, and analysis
for record in task_config["record"]:
record = record.copy()
if record["class"] == "SignalRecord":
srconf = {
"model": model,
"dataset": dataset,
"recorder": recorder
}
record["kwargs"].update(srconf)
sr = init_instance_by_config(record)
sr.generate()
else:
rconf = {"recorder": recorder}
record["kwargs"].update(rconf)
ar = init_instance_by_config(record)
ar.generate()
def begin_task_train(task_config: dict,
experiment_name: str,
recorder_name: str = None) -> Recorder:
"""
Begin task training to start a recorder and save the task config.
Args:
task_config (dict): the config of a task
experiment_name (str): the name of experiment
recorder_name (str): the given name will be the recorder name. None for using rid.
Returns:
Recorder: the model recorder
"""
with R.start(experiment_name=experiment_name, recorder_name=recorder_name):
R.log_params(**flatten_dict(task_config))
R.save_objects(**{"task": task_config
}) # keep the original format and datatype
R.set_tags(**{"hostname": socket.gethostname()})
recorder: Recorder = R.get_recorder()
return recorder
def end_task_train(rec: Recorder, experiment_name: str) -> Recorder:
"""
Finish task training with real model fitting and saving.
Args:
rec (Recorder): the recorder will be resumed
experiment_name (str): the name of experiment
Returns:
Recorder: the model recorder
"""
with R.start(experiment_name=experiment_name,
recorder_id=rec.info["id"],
resume=True):
task_config = R.load_object("task")
# model & dataset initiation
model: Model = init_instance_by_config(task_config["model"])
dataset: Dataset = init_instance_by_config(task_config["dataset"])
# model training
model.fit(dataset)
R.save_objects(**{"params.pkl": model})
# this dataset is saved for online inference. So the concrete data should not be dumped
dataset.config(dump_all=False, recursive=True)
R.save_objects(**{"dataset": dataset})
# generate records: prediction, backtest, and analysis
records = task_config.get("record", [])
if isinstance(records, dict): # prevent only one dict
records = [records]
for record in records:
cls, kwargs = get_cls_kwargs(
record, default_module="qlib.workflow.record_temp")
if cls is SignalRecord:
rconf = {"model": model, "dataset": dataset, "recorder": rec}
else:
rconf = {"recorder": rec}
r = cls(**kwargs, **rconf)
r.generate()
return rec
def task_train(task_config: dict, experiment_name):
"""
task based training
Parameters
----------
task_config : dict
A dict describes a task setting.
"""
# model initiaiton
model = init_instance_by_config(task_config["model"])
dataset = init_instance_by_config(task_config["dataset"])
# start exp
with R.start(experiment_name=experiment_name):
# train model
R.log_params(**flatten_dict(task_config))
model.fit(dataset)
recorder = R.get_recorder()
R.save_objects(**{"params.pkl": model})
# generate records: prediction, backtest, and analysis
for record in task_config["record"]:
if record["class"] == SignalRecord.__name__:
srconf = {
"model": model,
"dataset": dataset,
"recorder": recorder
}
record["kwargs"].update(srconf)
sr = init_instance_by_config(record)
sr.generate()
else:
rconf = {"recorder": recorder}
record["kwargs"].update(rconf)
ar = init_instance_by_config(record)
ar.generate()
"close_cost": 0.0015,
"min_cost": 5,
"return_order": True,
},
}
# model initialization
model = init_instance_by_config(task["model"])
dataset = init_instance_by_config(task["dataset"])
# NOTE: This line is optional
# It demonstrates that the dataset can be used standalone.
example_df = dataset.prepare("train")
print(example_df.head())
# start exp
with R.start(experiment_name="workflow"):
R.log_params(**flatten_dict(task))
model.fit(dataset)
R.save_objects(**{"params.pkl": model})
# prediction
recorder = R.get_recorder()
sr = SignalRecord(model, dataset, recorder)
sr.generate()
# backtest. If users want to use backtest based on their own prediction,
# please refer to https://qlib.readthedocs.io/en/latest/component/recorder.html#record-template.
par = PortAnaRecord(recorder, port_analysis_config)
par.generate()
def _log_task_info(task_config: dict):
R.log_params(**flatten_dict(task_config))
R.save_objects(**{"task":
task_config}) # keep the original format and datatype
R.set_tags(**{"hostname": socket.gethostname()})
def run_exp(
task_config,
dataset,
experiment_name,
recorder_name,
uri,
model_obj_name="model.pkl",
):
model = init_instance_by_config(task_config["model"])
model_fit_kwargs = dict(dataset=dataset)
# Let's start the experiment.
with R.start(
experiment_name=experiment_name,
recorder_name=recorder_name,
uri=uri,
resume=True,
):
# Setup log
recorder_root_dir = R.get_recorder().get_local_dir()
log_file = os.path.join(recorder_root_dir,
"{:}.log".format(experiment_name))
set_log_basic_config(log_file)
logger = get_module_logger("q.run_exp")
logger.info("task_config::\n{:}".format(
pprint.pformat(task_config, indent=2)))
logger.info("[{:}] - [{:}]: {:}".format(experiment_name, recorder_name,
uri))
logger.info("dataset={:}".format(dataset))
# Train model
try:
if hasattr(model, "to"): # Recoverable model
ori_device = model.device
model = R.load_object(model_obj_name)
model.to(ori_device)
else:
model = R.load_object(model_obj_name)
logger.info(
"[Find existing object from {:}]".format(model_obj_name))
except OSError:
R.log_params(**flatten_dict(update_gpu(task_config, None)))
if "save_path" in inspect.getfullargspec(model.fit).args:
model_fit_kwargs["save_path"] = os.path.join(
recorder_root_dir, "model.ckp")
elif "save_dir" in inspect.getfullargspec(model.fit).args:
model_fit_kwargs["save_dir"] = os.path.join(
recorder_root_dir, "model-ckps")
model.fit(**model_fit_kwargs)
# remove model to CPU for saving
if hasattr(model, "to"):
old_device = model.device
model.to("cpu")
R.save_objects(**{model_obj_name: model})
model.to(old_device)
else:
R.save_objects(**{model_obj_name: model})
except Exception as e:
raise ValueError("Something wrong: {:}".format(e))
# Get the recorder
recorder = R.get_recorder()
# Generate records: prediction, backtest, and analysis
for record in task_config["record"]:
record = deepcopy(record)
if record["class"] == "MultiSegRecord":
record["kwargs"] = dict(model=model,
dataset=dataset,
recorder=recorder)
sr = init_instance_by_config(record)
sr.generate(**record["generate_kwargs"])
elif record["class"] == "SignalRecord":
srconf = {
"model": model,
"dataset": dataset,
"recorder": recorder
}
record["kwargs"].update(srconf)
sr = init_instance_by_config(record)
sr.generate()
else:
rconf = {"recorder": recorder}
record["kwargs"].update(rconf)
ar = init_instance_by_config(record)
ar.generate()
def main(xargs):
dataset_config = {
"class": "DatasetH",
"module_path": "qlib.data.dataset",
"kwargs": {
"handler": {
"class": "Alpha360",
"module_path": "qlib.contrib.data.handler",
"kwargs": {
"start_time":
"2008-01-01",
"end_time":
"2020-08-01",
"fit_start_time":
"2008-01-01",
"fit_end_time":
"2014-12-31",
"instruments":
xargs.market,
"infer_processors": [
{
"class": "RobustZScoreNorm",
"kwargs": {
"fields_group": "feature",
"clip_outlier": True
}
},
{
"class": "Fillna",
"kwargs": {
"fields_group": "feature"
}
},
],
"learn_processors": [
{
"class": "DropnaLabel"
},
{
"class": "CSRankNorm",
"kwargs": {
"fields_group": "label"
}
},
],
"label": ["Ref($close, -2) / Ref($close, -1) - 1"],
},
},
"segments": {
"train": ("2008-01-01", "2014-12-31"),
"valid": ("2015-01-01", "2016-12-31"),
"test": ("2017-01-01", "2020-08-01"),
},
},
}
model_config = {
"class": "QuantTransformer",
"module_path": "trade_models",
"kwargs": {
"loss": "mse",
"GPU": "0",
"metric": "loss",
},
}
task = {"model": model_config, "dataset": dataset_config}
model = init_instance_by_config(model_config)
dataset = init_instance_by_config(dataset_config)
# start exp to train model
with R.start(experiment_name="train_tt_model"):
R.log_params(**flatten_dict(task))
model.fit(dataset)
R.save_objects(trained_model=model)
# prediction
recorder = R.get_recorder()
print(recorder)
sr = SignalRecord(model, dataset, recorder)
sr.generate()
# backtest. If users want to use backtest based on their own prediction,
# please refer to https://qlib.readthedocs.io/en/latest/component/recorder.html#record-template.
par = PortAnaRecord(recorder, port_analysis_config)
par.generate()
