For this project, we will design an MLP with $3$ hidden layers, each containing $16$ neurons. The input layer will have $4$ neurons $($one for each feature of the Iris dataset$),$ and the output layer will have $3$ neurons $($one for each class of the Iris dataset$).$