Understanding Narx Neural Network code| Matlab

Поділитися
Вставка
  • Опубліковано 19 вер 2024
  • This video provides a deep knowledge of the code for the NARX Neural network in Matlab, as well as what each part of the code represents.
    Previous videos:
    1. Concept of Narx Neural Network
    • Narx Neural Network | ...
    2. How to develop a neural network model || FFNN|| Code
    • How to develop a neura...
    3. Artificial Neural Network
    • What is an artificial ...
    4. Types of Artificial Neural Networks
    • What are the types of ...
    5. How to calculate RMSE, R2 and NSC in Matlab
    • How to calculate RMSE,...
    6. Normalization of data | code| Matlab
    • Normalization of data ... .
    code for prediction of Narx Neural network
    trainFcn = 'trainlm';
    inputDelays = 1:2;
    feedbackDelays = 1:2;
    hiddenLayerSize = 12;
    u=con2seq(train_in);
    v=con2seq(train_tr);
    w=con2seq(test_in);
    net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize);
    [Xs,Xi,Ai,Ts] = preparets(net,u,{},v);
    net.trainParam.max_fail = 1000;
    net.trainParam.min_grad=1e-7;
    net.trainParam.epochs=1000;
    net = train(net,Xs,Ts,Xi,Ai);
    [Y,Xf,Af] = net(Xs,Xi,Ai);
    perf = perform(net,Ts,Y)
    y=cell2mat(Y);
    Error1(train_tr(1,3:3305),y,'l')
    [netc,Xic,Aic] = closeloop(net,Xf,Af);
    Yc = netc(w,Xic,Aic);
    yc=cell2mat(Yc);
    Error1(test_tr,yc,'l')

КОМЕНТАРІ • 7

  • @joysanyal21
    @joysanyal21 9 місяців тому

    I have a question: In a previous video you demonstrated ways of creating the input and target data in various lags in Excell. Here, in this code if we specify the input delay as 1:5, do we need to prepare the input in Excell with lag 1. lag 2..lag 5 arranged in adjacent columns or Matlab is clever enough to generate that on its one from a simple time series?

  • @zuhalhamza3235
    @zuhalhamza3235 5 місяців тому

    Please I want to send me the procedure of Matlab in pdf in order to usese in my work

  • @wissalmasmoudi3780
    @wissalmasmoudi3780 2 роки тому +1

    Thank you very much for your effort. Actually i have a question if you don't minde.
    When i export the narx code from nnstart its give me this function(X=tonndata(P) ,T=tonndata(T))
    I don't know what is the point from this function?!
    Also i want to predict with new values,can you please explain to us how to predict a multistep ahead .
    Thanks again.

    • @myknowledgespace
      @myknowledgespace  2 роки тому +1

      I am currently making a video with similar features of what you want. You can expect it in 3-4 days

    • @wissalmasmoudi3780
      @wissalmasmoudi3780 2 роки тому +1

      @@myknowledgespace thank you very much

  • @mohamedzamil392
    @mohamedzamil392 2 роки тому

    Great work, Could you also please share your code, How can I get in touch with you?