Generates files batches from PSG data. — write_batches

Generates train batches from PSG data to be used by the `train_batches()` function.

write_batches_psg(
  records,
  events,
  batches_path = tempdir(),
  channels = c("C3-M2", "C4-M1", "O1-M2", "E1-M2", "E2-M1", "1-2"),
  resample = 70,
  padding = 1,
  batches_size = 1024,
  verbose = TRUE
)

Arguments

records: Character vector of EDF files paths to be included in the train batches.
events: List of events dataframes containing hypnograms corresponding to EDF records in `records` parameter.
batches_path: Character. Path where batches files will be saved.
channels: Character vector. Channels labels to include in the dataset.
resample: Integer. Sample rate to resample selected signals.
padding: Epochs added before and after each epoch.
batches_size: Number of epochs in each batch file.
verbose: Boolean, display status messages or not.

References

Chambon, S., Galtier, M., Arnal, P., Wainrib, G. and Gramfort, A. (2018) A Deep Learning Architecture for Temporal Sleep Stage Classification Using Multivariate and Multimodal Time Series. IEEE Trans. on Neural Systems and Rehabilitation Engineering 26:(758-769).