MATLAB Answers


question regarding ntstool, pollution mortality example dataset

Asked by John
on 31 Jul 2012
Latest activity Answered by Greg Heath
on 2 Nov 2016


I have a question regarding the ntstool (neural network time series).

There is an example dataset called pollution mortality that you can load into the GUI and learn how to train a network.

The following code is also in the GUI:

[X,T] = pollution_dataset;
net = narxnet(1:2,1:2,10);
[Xs,Xi,Ai,Ts] = preparets(net,X,{},T);
net = train(net,Xs,Ts,Xi,Ai);
Y = net(Xs,Xi,Ai)

The dataset has 219 input timestamps. But when you execute this code, the graph has 500 timesteps.

Could anybody tell me if code is indeed predicting observations in the future? Where does it specify how many time steps to look ahead?

Thank you



Log in to comment.

1 Answer

Answer by Greg Heath
on 2 Nov 2016
 Accepted Answer

 [X,T] = pollution_dataset;
net = narxnet(1:2,1:2,10);
[ Ic N ] = size(X) %[ 1 508]
[ Oc N ] = size(T))%[ 1 508]
x = cell2mat(X);
t = cell2mat(T);
[ I N ] = size(x) %[ 8 508]
[ O N ] = size(t))%[ 3 508]

So, where do you get 219?

 net = narxnet(1:2,1:2,10);

Therefore, the predictions are 2 timesteps ahead using 10 hidden layer nodes.


Log in to comment.

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today