harry1357931 / NeuralNetwork-ErrorBackPropagation

The program in this repository creates a 3 layer neural network with 1 hidden layer to find to find out a non linear function that best fits the input data. In this program, i implemented Error Back Propagation Algorithm to train the neural network and used Gradient descent to find out local minimum of the function. Then, i used 5 fold cross validation to find out average fitness of a model.