Neural Network initialization?

14 views (last 30 days)
Ameni JOUILA
Ameni JOUILA on 3 Feb 2018
Answered: Greg Heath on 4 Feb 2018
Hi, for the ones familiar with Neural Network , does this code initialize weights and biaises, or should I initialize them ? thank you ! n=5; net=feedforwardnet(n); net=configure(net,inputs,targets);

Answers (1)

Greg Heath
Greg Heath on 4 Feb 2018
MATLAB will automatically initialize unweighted nets before training.
MATLAB will continue to train a preweighted net
Therefore, if you are training a net multiple times in a for loop, you should use INIT or CONFIGURE at the top of the loop to initialize a new design.
I have posted zillions of examples of double for loop designs in both the NEWSGROUP (comp.soft-sys.matlab) and ANSWERS. Search
greg for i=1:Ntrials
Hope this helps
Thank you for formally accepting my answer
Greg

Categories

Find more on Deep Learning Toolbox in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!