For loop defining the network architecture

% Solve an Input-Output Fitting problem with a Neural Network
% Script generated by Neural Fitting app
% Created 26-Apr-2020 11:07:28
%
% This script assumes these variables are defined:
%
%   Einleger_Binaer_Alle_inv_sortiert - input data.
%   Auslenkung_Alle_inv - target data.

x = Einleger_Binaer_Alle_inv_Sortiert;
t = Auslenkung_Alle_inv;

rng('default');

% Choose a Training Function
% For a list of all training functions type: help nntrain
% 'trainlm' is usually fastest.
% 'trainbr' takes longer but may be better for challenging problems.
% 'trainscg' uses less memory. Suitable in low memory situations.
trainFcn = 'trainlm';  % nicht Levenberg-Marquardt backpropagation, da schneller

% Create a Fitting Network
% hiddenLayerSize = 10;
% net_hiddenlayersize6_sortiert = fitnet(hiddenLayerSize,trainFcn);

% For more hidden layers, layer construction
hiddenLayer1Size = 10; 
hiddenLayer2Size = 10;
net = fitnet([hiddenLayer1Size hiddenLayer2Size],  trainFcn);

Hi

Now I have a neural network with two hidden layers. i want to expand it for more hidden layers.

Can I use a for-loop for layer construction? And if it is possible, how can I write it?

Answer 1

[Bhavya Chopra]

I understand that you want to create a neural network with multiple hidden layers instantiated using a for loop. The fitnet function can be provided a vector of hidden layer sizes. Assuming the same size for each hidden layer:

trainFcn = 'trainlm';
n = 5;                     % Number of hidden layers
s = 10;                    % Size of each hidden layer
 
layerSizes = ones(1, n)*s  % Creating vector with layer sizes
net = fitnet(layerSizes, trainFcn);

Alternatively, if the layers have varying sizes, for instance in reducing powers of two:

trainFcn = 'trainlm';
n = 5;                     % Number of hidden layers
layerSizes = zeros(1,n);
 
for i = 1:n
    layerSizes(i) = 2^(n-i+1);
end
net = fitnet(layerSizes, trainFcn);

[Jingyuan Yao]

thanks for the inspiration!