braindecode.datasets.SleepPhysionetChallenge2018#

class braindecode.datasets.SleepPhysionetChallenge2018(subject_ids='training', path=None, load_eeg_only=True, preproc=None, n_jobs=1)[source]#

Physionet Challenge 2018 polysomnography dataset.

Sleep dataset from https://physionet.org/content/challenge-2018/1.0.0/. Contains overnight recordings from 1983 healthy subjects.

The total size is 266 GB, so make sure you have enough space before downloading.

See fetch_pc18_data for a more complete description.

Parameters:

  • subject_ids (list(int) | str | None) – (list of) int of subject(s) to be loaded. - If None, loads all subjects (both training and test sets [no label associated]). - If “training”, loads only the training set subjects. - If “test”, loads only the test set subjects, no label associated! - Otherwise, expects an iterable of subject IDs.

  • path (None | str) – Location of where to look for the PC18 data storing location. If None, the environment variable or config parameter MNE_DATASETS_PC18_PATH is used. If it doesn’t exist, the “~/mne_data” directory is used. If the dataset is not found under the given path, the data will be automatically downloaded to the specified folder.

  • load_eeg_only (bool) – If True, only load the EEG channels and discard the others (EOG, EMG, temperature, respiration) to avoid resampling the other signals.

  • preproc (list(Preprocessor) | None) – List of preprocessors to apply to each file individually. This way the data can e.g., be downsampled (temporally and spatially) to limit the memory usage of the entire Dataset object. This also enables applying preprocessing in parallel over the recordings.

  • n_jobs (int) – Number of parallel processes.