Practicing Neural Network with keras.
Build a neural network with one hidden layer , and draw a comparision of the mean of the mean squared errors, and standard deviation of the mean squared errors, for the cases with a) normalised data, b) increasing epochs. c) increasing the number of hidden layers.
Can demonstrate the effects of hyper parameter tuning and changing the architecture of the neural network.