K-means clustering for lower triangle of matrix

I am trying to perform k-means clustering on the lower triangle of a 52x52 matrix (m=tril(data,-1)). Basically, I want my final product to look like the graphic attatched below, with the shapes roughly deliniating the boundries of my clusters within my matrix.

I started off by trying to run the code given as an example in the k-means clustering documentation:

                X=tril(data,-1);
                [idx,C] = kmeans(X,3);
                x1 = min(X(:,1)):0.01:max(X(:,1));
                x2 = min(X(:,2)):0.01:max(X(:,2));
                [x1G,x2G] = meshgrid(x1,x2);
                XGrid = [x1G(:),x2G(:)]; % Defines a fine grid on the plot
                idx2Region = kmeans(XGrid,3,'MaxIter',1,'Start',C);

However, when I try to run this code I get the error message

"Error using kmeans (line 236)

The 'Start' matrix must have the same number of columns as X."

What do I need to do to fix this? And/or is there a better way of going about this? Thank you so much!

Answer 1

[Walter Roberson]

 X=tril(data,-1);

That is going to createa a 52 x 52 matix

[idx,C] = kmeans(X,3);

Each row of the matrix will be treated as the coordinates of a point. The output will be 52 rows of indices in idx (number of rows in X), and C will be a 3 x 52 matrix (number of clusters by number of columns in X)

XGrid = [x1G(:),x2G(:)]; % Defines a fine grid on the plot

That will only have two columns.