how can i enter more than one input in neural network??

how can i enter more than one input matrix in neural network??

i want to train my network to select the target for more than one input ??

[Greg Heath]

A clearer explanation is needed.

1. What is the physical problem?

2. Regression (fitnet) or classification (patternnet)?

3. What are your inputs and size of the input matrix

4. Ditto for outputs

5. A sample of your code would help.

Greg

Answer 1

[Greg Heath]

Use a two dimensional column vector input with dim 1 service values and dim 2 quality values the target is a 1 dimensional scalar with the corresponding BW value.For N cases

 [ 2 N ] = size(input)
 [ 1 N ] = size(target)

Hope this helps.

Thank you for formally accepting my answer.

Greg

[Greg Heath]

NO!

>> input=[ 5 9; 6 6; 7 8; 5 6; 9 5; 6 4 ];

target=[ 1; 2; 3; 5; 6; 7 ];

>> size(input)

ans =

     6     2

>> size(target)

ans =

     6     1

[osw]

Sorry to annoy you, but i didn't understand what you want to say

when i try to do the previous i have an error that is

Error using network/train Inputs and targets have different numbers of samples.

can you explain to me what you mean and how i enter my data ??

[Greg Heath]

The input matrix is an IxN dimensional matrix of N I-dimensional vectors.

The output matrix is an OxN dimensional matrix of N O-dimensional vectors.

[osw]

thank you for your help