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

Table 14

Final ensemble for 2DCNN architectures.

Architecture	Activation function	First kernel size	Second kernel size	RMSE	MAPE	IOA	R^2
CNN	Linear	30	2	18.7697	0.2295	0.8805	0.6221
		20	3
		40	3
		30	4
		50	6
	ReLU	40	3
		10	4
	Sigmoid	50	3
		50	4
	Softplus	10	5
		30	5
		20	6