Skip to main content
. 2023 Sep 7;16(18):6115. doi: 10.3390/ma16186115

Table 6.

Pseudocode of the ANN metamodel developed and investigated in this study.

PROCEDURE ARTIFICIAL NEURAL NETWORK
BEGIN
 Split dataset into training set and validation set
 Scale input values for the training set
 Scale the output values for ‘Vent_XY’ dataset
 Initialise parameter W and b
 Define the epoch, learning rate, and number of hidden neurons
   REPEAT
 Forward propagation
 Compute the loss using Root Mean Square Error
 Calculate the gradient of loss
 Calculate new W and b using the gradient and update the parameters
   UNTIL  Max epoch has reached
   RETURN  Trained model with updated parameters (W, b)
 Trained model is used to predict the test dataset
 RMSE of the model prediction and real value is presented
END