Fig. 3

A MLP training and validation accuracy for An. funestus age classes as training time increases (epoch; number of iterations over the entire dataset during the training process, i.e. seconds/iterations). Confusion matrix for predicting the age class of An. funestus. B The results for the MLP trained with important features/wavenumbers (n = 100) identified by XGBoost. C The results for the MLP method trained with eight principal components