Modify function to take input argument of marker

In this data cluster function the plot marker by default is '.'

https://in.mathworks.com/matlabcentral/fileexchange/31710-plotclusters-data-idx-centers-colors

I want to add a input argument to the function that will let me set the marker manually for separate inputs.

The modified function will be of the form

function []=PlotClusters(Data,IDX,Centers,point_marker,Colors)

Can someone please tell how to do this?

Answer 1

[Adam Danz]

I don't recommend editing the source code because 1) you'll have to repeate those edits every time the file is updated and 2) it's a lot more work than the first proposal below.

Proposal 1: change the marker type after plotting

Instead, just identify all of the points marked by '.' and replace the marker type.

PlotClusters(Data,IDX,Centers,Colors);         % run the code

axh = gca;                                     % get the axis handle
h = findobj(axh, 'Marker', '.');               % find all objects that have marker '.'
set(h, 'Marker', 'x')                          % change the marker type

Proposal 2: change the function

If you must edit the source code to specify the marker type, you'll need to make these changes:

% 1) Replace the first line
% function []=PlotClusters(Data,IDX,Centers,Colors)
  function []=PlotClusters(Data,IDX,Centers,Colors,Marker)

%2) Replace this line
% plot(Data(IDX == i,1),Data(IDX == i,2),'.','Color',Colors(i,:))
  plot(Data(IDX == i,1),Data(IDX == i,2),Marker,'Color',Colors(i,:))
  
%3) Replace this line
% plot3(Data(IDX == i,1),Data(IDX == i,2),Data(IDX == i,3),'.','Color',Colors(i,:)) 
  plot3(Data(IDX == i,1),Data(IDX == i,2),Data(IDX == i,3),Marker,'Color',Colors(i,:)) 
  
%4) Replace this line
% case 4 %All data is given just need to check consistency    
  case {4,5} %All data is given just need to check consistency  
      
%5 After the switch add these lines
if nargin < 5 || isempty(Marker)
    Marker = '.'; 
end

* Not tested!

** I recommend the first proposal!

[Meg Cullen]

I tried out both the proposals and Proposal 1 is actually much easier. Thank you very much. :)

[Adam Danz]

Good choice. The first one is definitely the way to go.