Changeset 66 for distools/protselfd.m


Ignore:
Timestamp:
07/08/13 18:17:33 (11 years ago)
Author:
bduin
Message:
 
File:
1 edited

Legend:

Unmodified
Added
Removed
  • distools/protselfd.m

    r30 r66  
    1515%
    1616% OUTPUT
    17 %   W     Selection mapping ('feature selection')
     17%   W     Selection mapping ('feature selection') or prototype indices.
    1818%   E     Error stimate as a function of number of selected prototypes
    1919%         (for supervised selection only reliable for prototype sizes >= class size)
     
    4646% In case of unsupervised selection the maximum or the mean distances to
    4747% the nearest prototype are minimized. These criteria are the same as used
    48 % in the KCENTRE and KMEDIOD cluster procedures.
     48% in the KCENTRE and KMEDIOD cluster procedures. What is returned now in W
     49% is the (ordered) list of prototype indices and not a mapping.
    4950%
    5051% REFERENCE
     
    6061% P.O. Box 5031, 2600 GA Delft, The Netherlands
    6162
    62 %
    63 
    6463function [R,e,D] = protselfd(D,ksel,type)
    6564
     
    7776      [R,e,D,J,nlab,clab] = protselfd(D,ksel,type);
    7877    case {'maxdist','meandist'}
    79       R = protselfd_unsuper(D,ksel,type);
     78      [R,e] = protselfd_unsuper(D,ksel,type);
    8079    otherwise
    8180      error('Unknown selection type')
     
    182181%   D     Square dissimilarity matrix, zeros on diagonal
    183182%   P     Number of prototypes to be selected
    184 %   CRIT  'dist' or 'centre'
     183%   CRIT  'maxdist' or 'meandist'
    185184%
    186185% OUTPUT
     
    197196%
    198197% SEE ALSO
    199 % KCENTRES
     198% KCENTRES, KMEDIODS
    200199
    201200% Copyright: R.P.W. Duin, r.p.w.duin@prtools.org
     
    203202% P.O. Box 5031, 2600 GA Delft, The Netherlands
    204203
    205 function N = protselfd_unsuper(d,p,crit)
     204function [N,e] = protselfd_unsuper(d,p,crit)
    206205
    207206d = +d;
Note: See TracChangeset for help on using the changeset viewer.