ValueError: CV (1) cannot be larger than number of subjects (0)

[kun2001github]

Hello Mr Perslev, When I tried to reproduce, I encountered the following error while running the data set segmentation：

(u-sleep) C:\Users\xxa\Desktop\U-Time-main\U-Time-main>ut cv_split --data_dir data/dcsm/ --subject_dir_pattern 'tp*'  --CV 1  --validation_fraction 0.10 --test_fraction 0.25  --seed 123

2024/04/07 19:06:09 | INFO | Entry script args dump: {'script': 'cv_split', 'project_dir': './', 'log_dir': 'logs', 'log_level': 'INFO', 'seed': 123}

2024/04/07 19:06:09 | INFO | Project directory set: C:\Users\xxa\Desktop\U-Time-main\U-Time-main (initialized project: False)

2024/04/07 19:06:10 | INFO | Seeding TensorFlow, numpy and random modules with seed: 123

2024/04/07 19:06:11 | INFO | Logs will not be saved to file for these loggers: [<Logger utime (INFO)>, <Logger psg_utils (INFO)>] (--log_file_name is empty)

2024/04/07 19:06:11 | INFO | Creating directory at C:\Users\xxa\Desktop\U-Time-main\U-Time-main\data\dcsm\views\fixed_split
Traceback (most recent call last):
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\Scripts\ut-script.py", line 33, in <module>
    sys.exit(load_entry_point('utime==1.1.8', 'console_scripts', 'ut')())
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\lib\site-packages\utime-1.1.8-py3.7.egg\utime\bin\ut.py", line 103, in entry_func
    mod.entry_func(script_args + help_agrs)
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\lib\site-packages\utime-1.1.8-py3.7.egg\utime\bin\cv_split.py", line 404, in entry_func
    run(args)
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\lib\site-packages\utime-1.1.8-py3.7.egg\utime\bin\cv_split.py", line 336, in run
    raise ValueError(f"CV ({n_splits}) cannot be larger than number of "
ValueError: CV (1) cannot be larger than number of subjects (0)
———————————————————————————————————————————————————————
(u-sleep) C:\Users\xxa\Desktop\U-Time-main\U-Time-main>ut cv_split --data_dir data/sedf_sc/  --subject_dir_pattern 'SC*'  --CV 1 --validation_fraction 0.10 --test_fraction 0.25 --subject_matching_regex 'SC4(\d{2}).*'   --seed 123

2024/04/07 19:05:57 | INFO | Entry script args dump: {'script': 'cv_split', 'project_dir': './', 'log_dir': 'logs', 'log_level': 'INFO', 'seed': 123}

2024/04/07 19:05:57 | INFO | Project directory set: C:\Users\xxa\Desktop\U-Time-main\U-Time-main (initialized project: False)

2024/04/07 19:05:58 | INFO | Seeding TensorFlow, numpy and random modules with seed: 123

2024/04/07 19:05:59 | INFO | Logs will not be saved to file for these loggers: [<Logger utime (INFO)>, <Logger psg_utils (INFO)>] (--log_file_name is empty)

2024/04/07 19:05:59 | INFO | Creating directory at C:\Users\xxa\Desktop\U-Time-main\U-Time-main\data\sedf_sc\views\fixed_split

2024/04/07 19:05:59 | INFO | OBS: Pairing files based on regex 'SC4(\d{2}).*'
Traceback (most recent call last):
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\Scripts\ut-script.py", line 33, in <module>
    sys.exit(load_entry_point('utime==1.1.8', 'console_scripts', 'ut')())
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\lib\site-packages\utime-1.1.8-py3.7.egg\utime\bin\ut.py", line 103, in entry_func
    mod.entry_func(script_args + help_agrs)
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\lib\site-packages\utime-1.1.8-py3.7.egg\utime\bin\cv_split.py", line 404, in entry_func
    run(args)
  File "H:\SoftWares\Working\Anaconda\envs\u-sleep\lib\site-packages\utime-1.1.8-py3.7.egg\utime\bin\cv_split.py", line 336, in run
    raise ValueError(f"CV ({n_splits}) cannot be larger than number of "
ValueError: CV (1) cannot be larger than number of subjects (0)

[happybegining]

Hello, @kun2001github, can you open and take a screenshot to see the folder of the dataset? I think this error occurs when your data set is empty.