Changeset 18 for distools/disnorm.m
- Timestamp:
- 06/15/11 15:05:46 (13 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
distools/disnorm.m
r16 r18 1 1 %DISNORM Normalization of a dissimilarity matrix 2 2 % 3 % V = disnorm(D,OPT)3 % V = DISNORM(D,OPT) 4 4 % F = E*V 5 5 % 6 6 % INPUT 7 7 % D NxN dissimilarity matrix or dataset, which sets the norm 8 % E Matrix to be normalized 8 % E Matrix to be normalized, e.g. D itself 9 9 % OPT 'max' : maximum dissimilarity is set to 1 by global rescaling 10 10 % 'mean': average dissimilarity is set to 1 by global rescaling (default) … … 17 17 % OPT = 'mean' 18 18 % 19 % DESCRIPTION 20 % Operation on dissimilarity matrices, like the computation of classifiers 21 % in dissimilarity space, may depend on the scaling of the dissimilarities 22 % (a single scalar for the entire matrix). This routine computes a scaling 23 % for a giving matrix, e.g. a training set and applies it to other 24 % matrices, e.g. the same training set or based on a test set. 19 25 20 26 % Copyright: Elzbieta Pekalska, ela.pekalska@googlemail.com … … 41 47 %DEFINE mapping 42 48 if isstr(opt) 43 discheck(D);49 % discheck(D); 44 50 opt = lower(opt); 45 51 if strcmp(opt,'mean')
Note: See TracChangeset
for help on using the changeset viewer.