/NeuralNet

A Simple Neural Network Engine in c++ which implements back-propagation algorithm it contains lots of flaws and it is intended to used for fun only

Primary LanguageC++MIT LicenseMIT

NeuralNet

It started in my junior high and now i want it to atleast behave like a neural net engine or do some stuff

any contributions ?? feel free to add them

#To train the network :-

INPUT LAYER HIDDEN LAYER OUTPUT LAYER
(N1,W1) (N1,W1)..(N1,W1)
(N2,W2) (N2,W2)..(N2,W2) OUTPUT
(N3,W3) .............
...... ............
...... (N,W)..(N,W)
(N,W)__ _________________________ ___________________________
OUTPUT = SIGMOID(N1W1 + N2W2 + .. + Nn*Wn)
Calculates error in the Weights from gradient descendant method
if desired output && desired error has been reached it stops updating the Weights and conclude that the Network has been trained much