KecNet: A Light Neural Network for Arrhythmia Classification Based on Knowledge Reinforcement

Table 1

Model parameters.

Layer		Feature map element no.	Kernel size	Stride
Sinc module	Sinc-Conv1D	32	251	1
	Max-pooling	—	2	2
	Dropout	—	—	—
	BN	—	—	—
Conv module	Conv1D	16	5	1
		16	5	1
	Max-pooling	—	2	2
	Dropout	—	—	—
	BN	—	—	—
GAP	16 + 1	—	—	—
Dense	16	—	—	—
Dropout	—	—	—	—
Dense	8		—	—
Softmax	5	—	—	—