Help with Training a Neural Network



Hi,

I'm trying to make neural network in matlab and then train it. I have
been reading the help that has come along with it but i cant seem to
make it work as i want it to. I am fairly new to the concept of neural
networks so i don't know how to do what i want to do correctly and i
keep getting errors.

The problem i have is that i have my input as an array of 1080
elements and 12 sets of these inputs (i.e. an array of 1080x12) and is
saved in a variable called 'Ar'. There is another similar set of
inputs that is just as big as 'Ar' (1080x12) which is stored in a
variable 'Br'.

I then make two target matrix of 26 elements using:

a=[1;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0];
target=[a,a,a,a,a,a,a,a,a,a,a,a];
b=[0;1;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0;0];
target2=[b,b,b,b,b,b,b,b,b,b,b,b];

This is so that each column in the variable target has the same data
as 'a' which represents the output based on one of the inputs (i.e. Ar
( : , 1 ) ) out of the 12 input sets in 'Ar'.

Next, following the help, i created a new neural network called 'net'
with 540 neurons in a hidden layer using:

net = newff (Ar , target , 540 );

Now i open up the neural network toolbox with the 'nntool' command. In
there i import the first data-set (Ar), the target data-set (target)
and the neural network created (net).

Next i open up the neural network and go to the training section when
i select the input and target data. I then train the network. I do the
same with the second data-set (Br) and its corresponding target data-
set (target2) and then train the network again.

Now when i run a simulation with any piece of data it always gives me
the same answer which is:

[1; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0;
0; 0; 0]

I don't know what I'm missing here and why the answer is always the
same.

I would really appreciate it if someone could point out where i have
made the mistake or what i have forgotten and guide me how to do it.
Thank you for any help you could provide.
.