Qlib RL 是否不支持一分鐘數據訓練嗎?

(qqlib) C:\Users\man-pc\qlib>python examples/rl_order_execution/scripts/gen_pickle_data.py -c examples/rl_order_execution/scripts/pickle_data_config.yml
[18380:MainThread](2024-04-23 08:22:19,538) INFO - qlib.Initialization - [config.py:416] - default_conf: client.
[18380:MainThread](2024-04-23 08:22:20,609) INFO - qlib.Initialization - [init.py:74] - qlib successfully initialized based on client settings.
[18380:MainThread](2024-04-23 08:22:20,609) INFO - qlib.Initialization - [init.py:76] - data_path={'1min': WindowsPath('C:/Users/man-pc/qlib/data/bin')}
[18380:MainThread](2024-04-23 08:22:20,610) INFO - qlib.HighFreqProvider - [highfreq_provider.py:144] - [qlib.contrib.data.highfreq_provider]Generating dataset
[18380:MainThread](2024-04-23 08:22:24,047) INFO - qlib.timer - [log.py:127] - Time cost: 2.972s | Loading data Done
C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\contrib\data\highfreq_processor.py:52: RuntimeWarning: Mean of empty slice
self.feature_mean = np.nanmean(df_values)
C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\numpy\lib\nanfunctions.py:1878: RuntimeWarning: Degrees of freedom <= 0 for slice.
var = nanvar(a, axis=axis, dtype=dtype, out=out, ddof=ddof,
[18380:MainThread](2024-04-23 08:22:24,051) ERROR - qlib.workflow - [utils.py:41] - An exception has been raised[ValueError: zero-size array to reduction operation fmax which has no identity].
File "examples/rl_order_execution/scripts/gen_pickle_data.py", line 30, in
feature = provider._gen_dataframe(deepcopy(provider.feature_conf))
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\contrib\data\highfreq_provider.py", line 147, in gen_dataframe
dataset = init_instance_by_config(config)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\utils\mod.py", line 174, in init_instance_by_config
return klass(**cls_kwargs, **try_kwargs, **kwargs)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset_init.py", line 119, in init
self.handler: DataHandler = init_instance_by_config(handler, accept_types=DataHandler)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\utils\mod.py", line 174, in init_instance_by_config
return klass(**cls_kwargs, **try_kwargs, **kwargs)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\contrib\data\highfreq_handler.py", line 134, in init
super().init(
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset\handler.py", line 468, in init
super().init(instruments, start_time, end_time, data_loader, **kwargs)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset\handler.py", line 100, in init
self.setup_data()
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset\handler.py", line 619, in setup_data
self.fit_process_data()
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset\handler.py", line 487, in fit_process_data
self.process_data(with_fit=True)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset\handler.py", line 551, in process_data
_infer_df = self._run_proc_l(_infer_df, self.infer_processors, with_fit=with_fit, check_for_infer=True)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\data\dataset\handler.py", line 498, in _run_proc_l
proc.fit(df)
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\qlib\contrib\data\highfreq_processor.py", line 58, in fit
np.save(self.feature_save_dir + name + "_vmax.npy", np.nanmax(df_values))
File "<array_function internals>", line 180, in nanmax
File "C:\Users\man-pc\AppData\Local\anaconda3\envs\qqlib\lib\site-packages\numpy\lib\nanfunctions.py", line 476, in nanmax
res = np.fmax.reduce(a, axis=axis, out=out, **kwargs)
ValueError: zero-size array to reduction operation fmax which has no identity

Start with examples/rl_order_execution/scripts/gen_pickle_data.py and pickle_data_config.yml, then trace the one-minute data path through qlib/contrib/data/highfreq_provider.py and highfreq_processor.py. Reproduce the command and determine why processing receives an empty array. Done means the one-minute dataset generation either completes successfully or reports the limitation clearly instead of raising the shown reduction error.