Prev: Time Difference
Next: Reading C code in matlab... mex file?
From: fenn lai on 8 Sep 2007 17:23
I tried to do prediction according to the data in data.txt
using the kprediction from I tried to do prediction
according to the data in data.txt using the kprediction from
your toolbox.

The code I used is :

************************************************************************************************************************************
load data.txt
A = data(:, 2);
Y = A';

PHI = 20*rand(1, 5790);
W1 = [ ];
W2 = [ ];
NetDef = ['HHLL'
'L---'];


trparms = settrain;
trparms=settrain(trparms,'maxiter',200);

[W1,W2,PI_vector,iter,lambda]=marq(NetDef,W1,W2,PHI,Y,trparms);

U = 27*rand(1, 5790);

NN = [2];

[W1,W2,PI_vector,iteration,lambda]=nnarx(NetDef,NN,W1,W2,trparms,Y,U);

Yk = kpredict('nnarx',NetDef,NN,10,W1,W2,Y,U)


************************************************************************************************************************************

but it shows error :

******************************************************************
Error using ==> plus
Matrix dimensions must agree.
Error in ==> kpredict at 83
nmax = max([na,nb+nk-1]); % 'Oldest' signal
used as input to the model

******************************************************************

when it's trying to execute the last command : Yk =
kpredict('nnarx',NetDef,NN,10,W1,W2,Y,U).

May you help me with this? thanks alot!
toolbox.
The code I used is :

************************************************************************************************************************************
load data.txt
A = data(:, 2);
Y = A';

PHI = 20*rand(1, 5790);
W1 = [ ];
W2 = [ ];
NetDef = ['HHLL'
'L---'];


trparms = settrain;
trparms=settrain(trparms,'maxiter',200);

[W1,W2,PI_vector,iter,lambda]=marq(NetDef,W1,W2,PHI,Y,trparms);

U = 27*rand(1, 5790);

NN = [2];

[W1,W2,PI_vector,iteration,lambda]=nnarx(NetDef,NN,W1,W2,trparms,Y,U);

Yk = kpredict('nnarx',NetDef,NN,10,W1,W2,Y,U)


************************************************************************************************************************************

but it shows error :

******************************************************************
Error using ==> plus
Matrix dimensions must agree.
Error in ==> kpredict at 83
nmax = max([na,nb+nk-1]); % 'Oldest' signal
used as input to the model

******************************************************************

when it's trying to execute the last command : Yk =
kpredict('nnarx',NetDef,NN,10,W1,W2,Y,U).

May help me with this? thanks alot!

 | 
Pages: 1
Prev: Time Difference
Next: Reading C code in matlab... mex file?