How to train and optimised my neural network

Question

I want to implement perceptron with python and ı use following equations, but how can ı set bias and how to update it.

You can click the links below to see the formulas.

where x is a D dimensional input vector, w is a D dimensional weight vector and b is a scalar

I use back propagation algorithm to train perceptron

score 0 · Answer 1 · answered Mar 27 '20 at 18:11

0

Maybe you should use Keras and create custom functions with it, it is a really user-friendly framework, here are some links to help you:

If you want to do it in plain Python (I wouldn't recommend), I would write all the derivatives to clarify my mind on how to do it.

answered Mar 27 '20 at 18:11

Bruno Mello

Could you please write derivatives? – Hakan Murat Aksüt Mar 27 '20 at 18:25
Do you want the derivative of the bias? – Bruno Mello Mar 27 '20 at 18:27
If you want a mathematical point of view of this question, it is better to ask it in https://math.stackexchange.com/. Stack overflow is more about implementation. – Bruno Mello Mar 27 '20 at 18:39

1 Answers1