selfeeg.dataloading

The dataloading module offers functionalities for data splitting with various desired settings, and for efficiently build PyTorch Dataloaders.

dataloading.load module

Classes

EEGDataset	custom pytorch.Dataset class that manages different loading configurations.
EEGSampler	custom pytorch Sampler designed to efficiently reduce the file loading operations.

Functions

get_eeg_partition_number	Calculates the number of unique partitions in each EEG signal.
get_eeg_split_table	creates a split table defining the files to use as train, validation and test sets.
get_eeg_split_table_kfold	creates a table with multiple splits for cross-validation.
check_split	check_split calculate and print split ratios to check if the split has been performed correctly.
get_split	extracts a split from the output of the get_eeg_split_table_kfold.