TypeError: cannot pickle '_thread.RLock' object

[dingtine]

[2628:MainThread](2021-09-29 10:48:13,029) ERROR - qlib.workflow - [utils.py:38] - An exception has been raised[TypeError: cannot pickle '_thread.RLock' object].
File "/Library/Frameworks/Python.framework/Versions/3.8/bin/qrun", line 8, in
sys.exit(run())
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/cli.py", line 62, in run
fire.Fire(workflow)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/fire/core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/fire/core.py", line 466, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/fire/core.py", line 681, in _CallAndUpdateTrace
component = fn(*varargs, kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/cli.py", line 56, in workflow
recorder = task_train(config.get("task"), experiment_name=experiment_name)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/model/trainer.py", line 101, in task_train
recorder = end_task_train(recorder, experiment_name)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/model/trainer.py", line 65, in end_task_train
R.save_objects({"params.pkl": model})
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/init.py", line 449, in save_objects
self.get_exp().get_recorder().save_objects(local_path, artifact_path, **kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/recorder.py", line 311, in save_objects
pickle.dump(data, f)
TypeError: cannot pickle '_thread.RLock' object

When run TFT sample model, got the above error.

---

**- Python version: 3.8

• OS : MacOS
• device: CPU**

---

[dingtine]

Run the TFTmodel, but use the cpu function.
My python version is 3.8.

Can change the keras store method from pickle to the kerasModel.save()?

[you-n-g]

We are fixing this bug.
You can try this temporary workaround #483

[dingtine]

We are fixing this bug. You can try this temporary workaround #483

when i update code to the new branch, a new error accur.

ERROR - qlib.workflow - [utils.py:38] - An exception has been raised[TypeError: expected str, bytes or os.PathLike object, not TFTModel].
File "/Library/Frameworks/Python.framework/Versions/3.8/bin/qrun", line 8, in
sys.exit(run())
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/cli.py", line 62, in run
fire.Fire(workflow)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/fire/core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/fire/core.py", line 466, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/fire/core.py", line 681, in _CallAndUpdateTrace
component = fn(*varargs, kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/cli.py", line 56, in workflow
recorder = task_train(config.get("task"), experiment_name=experiment_name)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/model/trainer.py", line 101, in task_train
recorder = end_task_train(recorder, experiment_name)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/model/trainer.py", line 65, in end_task_train
R.save_objects({"params.pkl": model})
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/init.py", line 449, in save_objects
self.get_exp().get_recorder().save_objects(local_path, artifact_path, **kwargs)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/workflow/recorder.py", line 312, in save_objects
Serializable.general_dump(data, path)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/utils/serial.py", line 140, in general_dump
obj.to_pickle(path)
File "/Users/tine/Downloads/qlib-fix_model_bug/examples/benchmarks/TFT/tft.py", line 315, in to_pickle
super().to_pickle(self, path / "qlib_model")
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/qlib/utils/serial.py", line 87, in to_pickle
with Path(path).open("wb") as f:
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/pathlib.py", line 1042, in new
self = cls._from_parts(args, init=False)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/pathlib.py", line 683, in _from_parts
drv, root, parts = self._parse_args(args)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/pathlib.py", line 667, in _parse_args
a = os.fspath(a)
TypeError: expected str, bytes or os.PathLike object, not TFTModel

[zhupr]

@dingtine
Line 314 of qlib/examples/benchmarks/TFT/tft.py is amended to : super(TFTModel, self).to_pickle(path / "qlib_model")

2efe95f

[you-n-g]

@dingtine
You can try TFT in the main branch.
It is updated and I successfully ran TFT several days ago.