Fig. 8

The loss and accuracy curves are presented for the training and validation sets, respectively, for φ = 0.2. A The loss function is a measure of the network's performance on the training set and gives an overview of the training process, indicating whether the network is on track. B The accuracy curve shows the training and validation accuracies. If there is a considerable difference between the two, it indicates overfitting, and the size of the gap provides an indication of how severe the overfitting is. The overall accuracy percentage achieved is 52.74%