newsom

Purpose

Create self-organizing map

Syntax

• net = newsom(P,[D1,D2,...],TFCN,DFCN,STEPS,IN)
  

Description

Competitive layers are used to solve classification problems.

net = newsom(P,[D1,D2,...],TFCN,DFCN,STEPS,IN) takes

P	R x Q matrix of Q representative input vectors.
Di	Size of ith layer dimension. Defaults = [5 8].
TFCN	Topology function. Default = 'hextop'.
DFCN	Distance function. Default = 'linkdist'.
STEPS	Steps for neighborhood to shrink to 1. Default = 100.
IN	Initial neighborhood size. default = 3.

and returns a new self-organizing map.

The topology function TFCN can be hextop, gridtop, or randtop. The distance function can be linkdist, dist, or mandist.

Properties

Self-organizing maps (SOM) consist of a single layer with the negdist weight function, netsum net input function, and the compet transfer function.

The layer has a weight from the input, but no bias. The weight is initialized with midpoint.

Adaption and training are done with trains and trainr, which both update the weight with learnsom.

Examples

The input vectors defined below are distributed over a two-dimensional input space varying over [0 2] and [0 1]. This data is used to train an SOM with dimensions [3 5].

• load simpleclass_dataset
  net = newsom(simpleclassInputs,[8 8]);
  plotsom(net.layers{1}.positions)
  net = train(net,simpleclassInputs);
  
  
  plot(simpleclassInputs(1,:),simpleclassInputs(2,:), ...
      '.g','markersize',20)
  hold on
  plotsom(net.iw{1,1},net.layers{1}.distances)
  hold off
  

See Also

sim, init, adapt, train, trains, trainr

Provide feedback about this page

newrbe

nftool

© 1984-2009- The MathWorks, Inc.    -   Site Help   -   Patents   -   Trademarks   -   Privacy Policy   -   Preventing Piracy   -   RSS