13.1.3 Unsupervised Network

Consider the case that you want to change the parameters in unsupervised networks.

Load the Neural Networks package.

In[1]:=<<NeuralNetworks`

The procedure is demonstrated on an unsupervised network with two inputs and four codebook vectors. Instead of submitting input data when initializing the network, enter a row matrix where the number of components equals the number of inputs.

Initialize an unsupervised network.

In[2]:=

Out[2]=

Retrieve information about the network.

In[3]:=

Out[3]=

The first component of the model is a list of four codebook vectors.

Look at the codebook vectors.

In[4]:=

Out[4]//MatrixForm=

There is one codebook vector on each row of the matrix. You may modify the values of the vectors by changing the corresponding elements of the matrix and inserting the modified matrix on the first position in the network.