An ANN model is designed to classifier datapoints into positive, negative, and neutral classes. There are $20$ input attributes for the model and $2$ hidden layers. The $1^{st}$ hidden layer has $10$ nodes, and the $2^{nd}$ hidden layer has $5$ nodes. How many parameters need to be trained in this model? $(5$ pt$)$

What technique do you need to evaluate the model in terms of whether the model is overfitting or underfitting? $(4$ pt$),$ briefly describe

What is overfitting? what is underfitting $?(6$ pt$)$