I\'m writing some basic neural network methods - specifically the activation functions - and have hit the limits of my rubbish knowledge of math. I understand the respective ran
Bipolar sigmoid = (1-e^(-x))/(1 + e^(-x))
Detailed explanation can be found at here