MATLAB: Error when using kmedoids with custom distance function

distance fucntionkmedoidsStatistics and Machine Learning Toolboxstatistics toolbox

Hi,

I created a custom distance function which I used with linkage and it worked fine. Here is how I call it: Z = linkage(matrix_clean,'average','@new_dist');

When I try using kmedoids (from the stats toolbox) with a custom fucntion I get an error. Here is how I call the fucntion: [idx,C,sumd,d,midx] = kmedoids(matrix_clean,2,'Distance','@new_dist');

Here is the error:

Error using lower Not enough input arguments.

Error in kmedoids (line 242) distance = lower(methods{i});

How should I pass in my custom distance function in kmedoids?

Thanks,

Daniel

Best Answer

Within the kmedoids function, there is a check on the distance metric to see if it is a string or a function handle. In your case, you have entered the function handle as a string and hence it is following a different code path from the one it should be following. All you would need to to is to use the command without quotes around your function handle:

[idx,C,sumd,d,midx] = kmedoids(matrix_clean,2,'Distance',@new_dist);

Related Solutions

MATLAB: Clustering Time Series with DTW

I think the error is due the reason that dtw function operates on 2 signals only and the output is always a scalar.

As per the description of 'Distance' Name Value Pair Argument of the kmedoids function:

"See pdist for the definition of each distance metric. kmedoids supports all distance metrics supported by pdist."

And as per the description of Distance input argument of the pdist function for custom distance:

"@distfun

Custom distance function handle. A distance function has the form

function D2 = distfun(ZI,ZJ)
% calculation of distance
...

where

ZI is a 1-by-n vector containing a single observation.
ZJ is an m2-by-n matrix containing multiple observations. distfun must accept a matrix ZJ with an arbitrary number of observations.
D2 is an m2-by-1 vector of distances, and D2(k) is the distance between observations ZI and ZJ(k,:)."

Hence you can't use the dtw function handle directly and you can use it as follows:

data = rand(161,20);
[idx,c,sumd,d] = kmedoids(data,10,'Distance',@dtwf);
function dist = dtwf(x,y)
% n = numel(x);
m2 = size(y,1);
dist = zeros(m2,1);
for i=1:m2
    dist(i) = dtw(x,y(i,:));
end
end

MATLAB: Error in using pdist2 , error in ExhaustiveSearcher/knnsearch

Your third party file c:\toolbox\classify\pdist2.m is interfering with use of the pdist2 from the Statistics toolbox. You need to rename or delete c:\toolbox\classify\pdist2.m or remove that directory from your MATLAB path.

Best Answer

Related Solutions

MATLAB: Clustering Time Series with DTW

MATLAB: Error in using pdist2 , error in Exhaustive​Searcher/k​nnsearch

Related Question

MATLAB: Error in using pdist2 , error in ExhaustiveSearcher/knnsearch