neural-nets

A repo for code created to explore what neural networks are, how they work, and how to build one. Created 9/2/2017

nnv2.py was created based on the linear perceptron neural network model created by Frank Rosenblatt. It is a supervised learning model which has 4 main steps:

1. Make a guess for the connection weights
2. Calculate the real output and compare it against an expected output
3. Use the error term to update the weights
4. Loop over a dataset, or until the error does not get smaller