Handwritten digits $28\backslash$times $28$ greyscale images. Features are the $784-$pixel greyscale values $(0,255).$ Labels are the class of digits $09.$ We want to build two hidden$-$layer networks with $200$ units in the first hidden layer, $100$ units in the second layer, and $10$ units in the output layers. How many neural weights in total are there in the multi$-$layer perceptron networks $($to simplify the calculation, you do not need to include the bias parameters$)?$