This project is an implementation of this resource https://blog.paperspace.com/constructing-neural-networks-from-scratch/
An simple neural net predicting the output given a pair of binary inputs. The function is XOR, which is not linearly separable, so a neural net is a good choice for modeling it.
We use a simple 2-layer neural network (with one hidden layer and one output layer), which is trained using a basic form of gradient descent.
The loss function is cross entropy.# xor-neural-net