Intra- and inter-epoch temporal context network (IITNet) using sub-epoch features for automatic sleep scoring on raw single-channel EEG.

•IITNet extracts representative features at a sub-epoch level from raw single-channel EEG.•Intra- and inter-epoch temporal contexts are captured in multiple successive epochs.•Influence of the sequence length on performance was analyzed on SleepEDF, MASS, and SHHS.•Intra-epoch temporal context learning with a residual network enhanced sleep scoring performance.•Using last two-minute epochs can be reasonable for efficient and reliable automatic sleep scoring.