Issue while running uSleep

[JBurchert]

Hello Mr. Perslev,

I want to run uSleep for the DCSM dataset. I have followed the provided instructions and ran ut fetch, extract, extract_hypno and cv_split for the given dataset, initialized the project, and modified the dataset configuration yaml. However when I want to train the model with ut train --num_gpus=1 --channels 'EEG Fpz-Cz' --overwrite --datasets dcsm I get the error below. Did I miss a step while setting up the dataset or do I need to add additional information for the appropriate regex somewhere?

I really appreciate any help you can provide.

Traceback (most recent call last): File "/home/burchert/miniconda3/envs/u-sleep/bin/ut", line 10, in sys.exit(entry_func()) File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/utime/bin/ut.py", line 103, in entry_func mod.entry_func(script_args + help_agrs) File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/utime/bin/train.py", line 306, in entry_func run(args=args) File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/utime/bin/train.py", line 216, in run train_datasets, val_datasets = dataset_func(hparams, args.no_val, args.train_on_val, args.datasets) File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/utime/utils/scriptutils/train.py", line 74, in get_train_and_val_datasets datasets = [*get_splits_from_all_datasets(hparams, load, dataset_ids=dataset_ids)] File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/utime/utils/scriptutils/scriptutils.py", line 221, in get_splits_from_all_datasets ds = get_dataset_splits_from_hparams( File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/utime/utils/scriptutils/scriptutils.py", line 166, in get_dataset_splits_from_hparams dataset = SleepStudyDataset(**hparams[data_key], File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/psg_utils/dataset/sleep_study_dataset/sleep_study_dataset.py", line 48, in init super(SleepStudyDataset, self).init( File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/psg_utils/dataset/sleep_study_dataset/subject_dir_sleep_study_dataset_base.py", line 72, in init ss = sleep_study_class( File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/psg_utils/dataset/sleep_study/sleep_study.py", line 76, in init super(SleepStudy, self).init( File "/home/burchert/miniconda3/envs/u-sleep/lib/python3.10/site-packages/psg_utils/dataset/sleep_study/subject_dir_sleep_study_base.py", line 67, in init raise ValueError("Could not uniquely infer PSG/HYP files in subject" ValueError: Could not uniquely infer PSG/HYP files in subject directory '/home/burchert/uSleep/datasets/processed/dcsm/views/fixed_split/train/tp7f6c419b_c612_40fb_8c83_1e63f5cd3f3d'. Consider specifying/correcting one or both of the psg_regex and hyp_regex parameters to explicitly select the appropriate file(s) within the subject dir. srun: error: test-000: task 0: Exited with exit code 1