Evaluation of Key Parameters Using Deep Convolutional Neural Networks for Airborne Pollution (PM10) Prediction

Table 6

Summary of 24 1DCNN with one convolutional layer architectures.

Architecture	Activation function	Mean
		RMSE	MAPE	IOA	R^2
CNN	Linear	19.4408	0.2375	0.8698	0.6
	ReLU	22.3676	0.2702	0.8297	0.4703
	Sigmoid	21.0785	0.2626	0.8575	0.529
	Softplus	21.0008	0.2574	0.857	0.5325
Total		20.9719	0.2569	0.8535	0.533