How many layers are hidden ? Is it Backpropagation network ?

How many layers are hidden ? Is it Backpropagation network ?

net = newff(minmax(alphabet),[S1 10 S2],{'logsig' 'logsig' 'logsig' },'traingdx');

net.LW{2,1} = net.LW{2,1}*0.01;

net.b{2} = net.b{2}*0.01;

net.performFcn = 'mse';

net.trainParam.show = 20;

net.trainParam.epochs = 2000;

net.trainParam.mc = 0.95;

P = alphabet;

T = targets;

[net,tr] = train(net,P,T);

Answer 1

[Jayanti]

Hi Silva,

Yes, this defines a feedforward backpropagation neural network. The training function used here is "traingdx", which stands for Gradient Descent with momentum and an adaptive learning rate a commonly used backpropagation training algorithm.

But the function "newff" has been obselete. You can refer to the below MATLAB Answers thread for more information:

https://in.mathworks.com/matlabcentral/answers/130820-understanding-the-newff-and-train-functions#answer_138080