I have designed a Neural Network model and plotted the loss curves on two different datasets. With more iterations, the gap between validation and train loss grows, but the AUC score of validation increases. Is this model acceptable? If not, how can I improve it?
