A really simple neural network which has been coded from scratch (without using libraries like keras/tensorflow etc.). The code goes through what a one layer neural network does. It takes in a two-dimensional vector input, and outputs a number in [0,1]. Currently, the neural net is trained so that vectors with modulus/norm greater than or equal to 3.5 output 1 and otherwise output 0 - although it is easy enough to change this training data around for another similarly simple task. A graph for the error of training is also generated.