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

Table 3

Summary of 40 MLP architectures.

Architecture	Activation Function	Mean
		RMSE	MAPE	IOA	R^2
MLP	Linear	19.764	0.2369	0.8695	0.5866
	ReLU	20.3885	0.2461	0.8576	0.56
	Sigmoid	20.3289	0.2493	0.8638	0.5626
	Softplus	20.105	0.2408	0.8608	0.5722
Total		20.1466	0.2433	0.8629	0.5703