This model contains a implementation of the SOFM algorithm using Simulink's basic blocks. The SOFM algorithm is associated with a single block with various configuration parameters:
Number of the neuron inputs
Grid size (rows and columns)
Initial value of standard deviation (sigma0) - Topological neighborhood function
Time constant (t1) - Topological neighborhood function decrease
Initial value of the learning-rate parameter (mu0)
Time constant (t2) - Learning-rate parameter decrease
The attached file contains an example of a network with two dimensional lattice driven by a two dimensional distribution with 100 neurons arranged in a 2D lattice of 10 x 10 nodes.
Marcelo Augusto Costa Fernandes
DCA - CT - UFRN
weights2DView.m (s-function) file was inserted.
Insert more information of the model in description field.