Using (classificationKNN.fit) in Matlab

Hi,
I want to define a set of weights for the distance measure in KNN method. According to the MATLAB help the distance weight function can be a 'equal','inverse','squaredinverse', or a function handle. I tried to define the weights using the following function.
function WGT=gene
weights=round(rand(a,b))
WGT=T
end
However, when I apply KNN the weights does not changed and are equally distributed. My fitting code is
(fitness=ClassificationKNN.fit(train_attribute,train_class,'Numneighbors',4,'distanceweight',@gene_weights))
Can anyone help with this problem please. Thank you in advance Best Regards

Answers (0)

Asked:

on 21 Oct 2013

Edited:

on 23 Oct 2013

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!