MATLAB: How to calculate the output error for feedforwardnet

Question

i have design my feedforwardnet according to the examples. but how do i calculate the misclassified output? this is where my program stop so far,

output=sim(net,inp);

i'm using -1 and 1 for my target

hope for some help since neural network is still something new to me. thank you

Answer 1

For classification of c classes, the target columns should be columns of the c-dimensional unit matrix.

class = vec2ind(target);

N = length(class);

....

assignedclass = vec2ind(output);

Nerr = sum(assignedclass ~= class);

PCTerr = 100*Nerr/N

Hope this helps.

Thank you for formally accepting my answer.

Greg