Table 3.

Complexity and resource metrics for the evaluated deep learning (DL) architectures.

DL architecture	Total parameters	Computational complexity (GFLOPSa)	Model size (MB)
CNNb 1D	43,366	2.41	0.27
CNN 2D	1,208,578	20.92	4.88
Attention model	407,620	80.25	2.68
ConvLSTMc	997,890	31.42	4.03

^aGFLOPS: giga floating-point operations per second.

^bCNN: convolutional neural network.

^cConvLSTM: convolutional long short-term memory.