In pattern recognition using neural network what should be the output?

Suppose I want to recognise a particular face I am extracting sift points from it and feeding it as a input vector . Then what should be the output ? I am talking during the training phase .

How are outputs decided for a particular type of input pattern?

Answer 1

[Greg Heath]

% For c classification categories numbered 1:c, use a classindex row vector

 classind = [ 5 3 1 4 2  5 4 3 2 1 ]
 N  = length(classind)

% The corresponding target matrix is obtained using the ind2vec command

 target = full(ind2vec(classind))

% The corresponding output will be a matrix of the same size

 outind = vec2ind(output)

 err = outind ~= classind

 Nerr = sum(err)

 PctErr = 100*Nerr/N

%Obviously, more is needed to obtain error rates for individual classes.

A search of both the NEWSGROUP and ANSWERS using

 greg patternnet

should yield details.

Hope this helps.

Thank you for formally accepting my answer.

Greg

[Newman]

Dear Sir, does c classification categories means c number of faces of c number of handwriting characters ?

[Greg Heath]

Whatever you want to classify.

 The face is that of Karen

or

the face has a scar under the left eye?

Greg