表 2. Average results comparison of automatic detection of arousal events based on the convolutional self-attention model and the models in references [14, 25-26] （）.

基于卷积注意力网络和文献[14, 25-26]中网络模型的觉醒事件自动检测平均结果（）

网络模型	准确率（%）	F1分数（%）	AUPRC（%）	AUROC（%）
卷积注意力网络	81.9 ± 5.0	74.9 ± 12.6	68.1 ± 17.2	78.9 ± 6.5
DRCNN	80.5 ± 5.2	69.4 ± 13.0	62.5 ± 19.8	65.1 ± 8.2
DeepSleep 2.0	75.3 ± 6.7	68.0 ± 13.3	60.1 ± 21.4	63.4 ± 11.7
DNNs	78.6 ± 6.1	72.6 ± 11.4	65.3 ± 19.6	70.2 ± 9.5