Neural Network Toolbox

nnt2som

Update NNT 2.0 self-organizing map

Syntax

net = nnt2som(PR,[D1,D2,...],W,OLR,OSTEPS,TLR,TND)

Description

• nnt2som(PR,[D1,D2,...],W,OLR,OSTEPS,TLR,TND) takes these arguments,

  PR -- R x 2 matrix of min and max values for R input elements

  Di -- Size of ith layer dimension

  W -- S x R weight matrix

  OLR -- Ordering phase learning rate, default = 0.9

  OSTEPS -- Ordering phase steps, default = 1000

  TLR -- Tuning phase learning rate, default = 0.02

  TND -- Tuning phase neighborhood distance, default = 1

and returns a self-organizing map.

nnt2som assumes that the self-organizing map has a grid topology (gridtop) using link distances (linkdist). This corresponds with the neighborhood function in NNT 2.0.

The new network will only output 1 for the neuron with the greatest net input. In NNT 2.0 the network would also output 0.5 for that neuron's neighbors.

Once a network has been updated, it can be simulated, initialized, adapted, or trained with sim, init, adapt, and train.

See Also

newsom

nnt2rb

nntool

© 1994-2005 The MathWorks, Inc.