SOM Toolbox | Online documentation | http://www.cis.hut.fi/projects/somtoolbox/ |
[sM,sTrain] = som_prototrain(sM, D)
SOM_PROTOTRAIN Use sequential algorithm to train the Self-Organizing Map. [sM,sT] = som_prototrain(sM, D) sM = som_prototrain(sM,D); Input and output arguments: sM (struct) map struct, the trained and updated map is returned (matrix) codebook matrix of a self-organizing map size munits x dim or msize(1) x ... x msize(k) x dim The trained map codebook is returned. D (struct) training data; data struct (matrix) training data, size dlen x dim This function is otherwise just like SOM_SEQTRAIN except that the implementation of the sequential training algorithm is very straightforward (and slower). This should make it easy for you to modify the algorithm, if you want to. For help on input and output parameters, try 'type som_prototrain' or check out the help for SOM_SEQTRAIN. See also SOM_SEQTRAIN, SOM_BATCHTRAIN.