Prev: Matlab 2010a Slow Editor
Next: 2D gaussian fitting using Surface Fitting Tool
From: Nehal on 9 Aug 2010 16:49
I have some variables.. I am putting the dimensions below...

P <672x1500 logical>
Tc <1x1500 double>
(Tc do not have any 0's)

now when I want to run PNN with these variables,

T = ind2vec(Tc); % this line works fine...
it shows a result of T <15x1500 double>

but...
net = newpnn(P,T); % this line shows error showed below...

??? Error using ==> network.subsasgn at 551
net.IW{1,1} must be a 1500-by-672 matrix.

Error in ==> newpnn at 103
net.iw{1,1} = p';

What am I doing wrong..? any help...?
From: Greg Heath on 11 Aug 2010 17:57
On Aug 9, 4:49 pm, "Nehal " <arnab...(a)yahoo.com> wrote:
> I have some variables.. I am putting the dimensions below...
>
> P <672x1500 logical>
> Tc <1x1500 double>
> (Tc do not have any 0's)
>
> now when I want to run PNN with these variables,
>
> T = ind2vec(Tc); % this line works fine...
> it shows a result of  T <15x1500 double>
>
> but...
> net = newpnn(P,T); % this line shows error showed below...
>
> ??? Error using ==> network.subsasgn at 551
> net.IW{1,1} must be a 1500-by-672 matrix.
>
> Error in ==> newpnn at 103
> net.iw{1,1} = p';
>
> What am I doing wrong..? any help...?

For I input nodes and O output nodes

size(P) = [I N]
size(T) = [O N]

Hope this helps.

Greg
 | 
Pages: 1
Prev: Matlab 2010a Slow Editor
Next: 2D gaussian fitting using Surface Fitting Tool