barkais adding network

learning rate = (this is how much the network is changed over each training)

problem: + =

output: