Machine Problem

Download the Deep Learning Toolbox.

Train a feed$-$forward backpropagation network using the whole real estate valuation dataset. Use all$/$some of the features:

For the purposes of this machine problem, try to aim for an accuracy of at least $85\%$ across the training, validation, testing, and overall results.

You are free to choose which features to use. You may try performing feature normalization.

Play around with the hyperparameters to get a high accuracy. $($no$.$ of layers, no of neurons per layer, and training function$)$

Store the trained network in a network object called myNetwork.

Save the script file as exp$6\_$idno.m$.$

For reproducibility of results, set a fixed rng seed using $"$rng$(0)"$