Neural Network Tools To Predicte Future Values

Hello all
I have decided to explore the Neural Network Tools available in MATLAB (despite having no experience or knowledge – best way I guess).
I was hoping someone could help point me in the right direction.
What I am trying to do is predict a set of future values based on a trained model of historic values, my data is shown below:-
NN Data.JPG
I have a set of values that was collected from 01-02-2019 to 10-08-2019, based on these values I wanted the neural network to predict the values for the 01-09-2019.
My first question is can anyone tell me how many layers should there be - or how does one determine this?
Also I know that by selecting a high value for the neurons takes longer but is more accurate and a low number for the neurons is quicker but less accurate – but can anyone help me as to how I can decide on the number of neurons I should select?
Any anyone provide any support.
Thank you.
(Currently using Matlab R2018a)

Answers (0)

This question is closed.

Asked:

on 12 Aug 2019

Closed:

on 20 Aug 2021

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!