Network parameters in the MATLAB software toolbox.

Topology	four inputs, one output, and one hidden layer with four neurons (4 × 4 × 1)
Data	Training set: 69.44% randomly selected observation data (50 data values)
	Test set: 15.27% randomly selected observation data (11 data values)
	Validation set: 15.27% randomly selected observation data (11 data values)
Beginning function	Log-sigmoid
Training algorithm	Levenberge–Marquardt
Loss function conditions	Minimum MSE
Stopping conditions	The network stops in one of three ways
	Validation check > 10
	Minimum gradient < 10−7
	Momentum speed > 1010