Hi, can anybody help with some guidance on neural network for classification ?

I'm working on optical character recognition problem. I've successfully extracted features which is a [1X32] matrix ('ve extracted 32 features from each segmented character). I've the complete training data set (the images of every individual character), but the problem is that I am not getting how to create Input & Target data set matrices.

Any help would be greatly appreciated.

Answer 1

[Greg Heath]

In general

 [ I N ] = size(input)  % I = 32
 [ O N ] = size(target) % O = 36

where the columns of target are columns of eye(36).

go to

 help nndatasets
 doc nndatasets

to see how the matrices of the classification examples are set up.

Hope this helps

Thank you for formally accepting my answers

Greg

[Nawaz Pasha]

Mr. Heath thank you for your answer.

I've few more simple questions since i'm a newbie to matlab hope you would be helpful.

1)There are 258 different patterns (characters), so, should there be 258 class labels ?

My input matrix size is No. of rows = 32 (features) No. of cols = 258*4=1032 (No of characters*No of instances for each character)

2) what should be the size of my target matrix ? Just draw a dummy target matrix for my case.

Thanks in advance..

[Greg Heath]

I don't understand the 258. 36 is the typical count for English 26(A-Z)+10(0-9) or perhaps 62 if lower case (a-z) is allowed. Well, even more if punctuation marks are included.