What is wrong on these code ?

Dear brother

I applied these code for forecasting 10 days head

I used NARNET approach but i got error a

also i am using matalb R2013a

These error

Index exceeds matrix dimensions.

t=load('D:\myexp\RKMS with logfun\2009data.csv');
time = 1:length(t)
n=length(t);

%-------------------------------------------------------------

%-----------------------------------------------------------------

t = tonndata(t,false,false);

%------------------------------------------------------------
% ---------------------------------DESIGN NETWORK
ID=1:2;
HL=20;
net = narnet(ID,HL);
%--------------------------------Training the network
[Xs,Xi,Ai,Ts] = preparets(net,{},{},t);
net = train(net,Xs,Ts,Xi,Ai);

%------------------------------------ SIMULATE NETWORK FOR ORIGINAL SERIES
[Y,Xf,Af] = sim(net,Xs,Xi,Ai);

%-----------------------------

   netc = closeloop(net);
  view(netc)
  % 10 More Steps for the first (now only) input
  X2 = num2cell(rand(1,10));

% Initial input states for closed loop continuation will be the
% first input's final states.
Xi2 = Xf(1,:);

% Initial 2nd layer states for closed loop contination will be the
% processed second input's final states.  Initial 1st layer states
% will be zeros, as they have no delays associated with them.
Ai2 = cell2mat(Xf(2,:));
for i=1:length(net.inputs{1}.processFcns)
  fcn = net.inputs{i}.processFcns{i};
  settings = net.inputs{i}.processSettings{i};
  Ai2 = feval(fcn,'apply',Ai2,settings);
end
Ai2 = mat2cell([zeros(10,2); Ai2],[10 1],ones(1,2));

% Closed loop simulation on X2 continues from open loop state after X.
Y2 = sim(netc,X2,Xi2,Ai2);

[Geoff Hayes]

abduall - which line of code generates this error? Please copy and paste the full error message into question.

Typically, the Index exceeds matrix dimensions. message indicates that your code is trying to access an array using an index that exceeds the dimensions of this array. For example,

 z = zeros(1,10);
 z(11) = 42;

would generate the same error that you are observing (since z has only ten elements and the second line of code is trying to access the 11th element).

[abduall hassn]

This part

Ai2 = cell2mat(Xf(2,:));
for i=1:length(net.inputs{1}.processFcns)
  fcn = net.inputs{i}.processFcns{i};
  settings = net.inputs{i}.processSettings{i};
  Ai2 = feval(fcn,'apply',Ai2,settings);
end
Ai2 = mat2cell([zeros(10,2); Ai2],[10 1],ones(1,2));

Answer 1

[Greig]

I suspect the problem is in the for loop, most probably here...

fcn = net.inputs{i}.processFcns{i};
settings = net.inputs{i}.processSettings{i};

"i" is indexing the elements of net.inputs.processFcns, not net.inputs Should these lines perhaps be the following?

fcn = net.inputs{1}.processFcns{i};
settings = net.inputs{1}.processSettings{i};

As a side, it is good practice not to use "i" or "j" for indexing in MATLAB. These are reserved for complex numbers. I tend to use "ii" or "jj".

[abduall hassn]

Dear Greig

i have done changing but stile same problem

[Walter Roberson]

Please show your current code and indicate which line the error message is appearing on.

[abduall hassn]

Walter Roberson sir the code showing above u can see it

this part show error

Ai2 = cell2mat(Xf(2,:));
for i=1:length(net.inputs{1}.processFcns)
  fcn = net.inputs{i}.processFcns{i};
  settings = net.inputs{i}.processSettings{i};
  Ai2 = feval(fcn,'apply',Ai2,settings);
end
Ai2 = mat2cell([zeros(10,2); Ai2],[10 1],ones(1,2));

[Greig]

To echo Walter, exactly what line is causing the problem? Not what part, the exact line.

What are the values for...

length(net.inputs{1}.processFcns)
length(net.inputs{1}.processFcns{1})

length(net.inputs{1}.processSettings)
length(net.inputs{1}.processSettings{1})

Is there a mismatch?

[Walter Roberson]

Ai2 = cell2mat(Xf(2,:));
for i=1:length(net.inputs{1}.processFcns)
  fcn = net.inputs{1}.processFcns{i};
  settings = net.inputs{1}.processSettings{i};
  Ai2 = feval(fcn,'apply',Ai2,settings);
end
Ai2 = mat2cell([zeros(10,2); Ai2],[10 1],ones(1,2));

[abduall hassn]

Sir Walter Robesron and Greg

When i did changed but still i have same problem

[Walter Roberson]

Show your exact changed code, and show the complete error message including the part where it shows exactly which line the problem is occurring on.

[abduall hassn]

for i=1:length(net.inputs{1}.processFcns)
  fcn = net.inputs{1}.processFcns{i};
  settings = net.inputs{1}.processSettings{i};
  Ai2 = feval(fcn,'apply',Ai2,settings);
end

Index exceeds matrix dimensions.

[Walter Roberson]

... and show the complete error message including the part where it shows exactly which line the problem is occurring on

Also at the command line give the command

dbstop if error

and run the code. When it stops with the error, show us

size(net.inputs{1}.processFcns)
size(net.inputs{1}.processSettings)

[abduall hassn]

Thank for ur answer

the previous error it was

 Ai2 = cell2mat(Xf(2,:));

but i solved it

but i got this error

 Number of input states does not match net.numInputs.

 Y2 = sim(netc,X2,Xi2,Ai2);

Thank u alot sir

[Greig]

I guess my suspicions about where the error was were wrong... Clearly, Sherlock Holmes I am not!

Now that the my powers of deductions are proven to be utterly useless, you will need to give us much more information about this new error.

The function "sim" is completely unknown to me and the error is internal to that function, so with out the code and exactly where the error is happening I'm afraid to say I, and probably most others, won't be of much help.

[Walter Roberson]

sim() is from the Neural Network toolbox

[abduall hassn]

Walter Roberson sir

I am very sorry for disturbed u

sir what happened that i am having 7 year data from 2008 to 2014 and this contains two attributes one date and either packets numbers

i want to forecast next period 2015

sir i got this code but i am getting error

can u provide me code for forecasting next period 2015

Thank so much sir

[Greig]

Walter - Thanks for the info, it is not a toolbox that I have.

Abduall - Try using size() on the inputs and then have a look at this answer HERE, where someone encounters a similar problem.