Table 8.

The results of discussion about temporal convolution networks. “M” means $\times 10^6$. Batch time means the runtime of each batch in the model testing.

Method	Accuracy	Precision	Sensitivity	Specificity	F1-score	AUC	The model size of temporal convolution networks	Batch time
Ours (TCN [72])	0.8679	0.8688	0.8641	0.8909	0.8664	0.8675	28.6M	1.77 s
Ours (TrellisNet [109])	0.8788	0.8824	0.8761	0.9036	0.8792	0.8842	87M	1.69 s
Ours (SA-TCN [110])	0.8854	0.8995	0.8837	0.9087	0.9040	0.8869	54M	1.31 s
Ours (TCAN [111])	0.8870	0.9302	0.8981	0.9248	0.9275	0.8895	33M	1.26 s
Ours	0.9810	0.9889	0.9861	0.9859	0.9875	0.9908	32M	1.14 s