|
Specification of Network dimensions |
Input nodes:
Each node in the input layer represents an input feature from the data set.
Dimensions:
You can specify the map to be 2 or 3 dimensional.
Once you have specified the dimensionality of the map, you can specify the size of each dimension.
|
Creating the network |
Once you have specified your network architecture press the Create button, or select Create Network from the Network menu.
This creates a network file with randomly initialised connection weights.
To retain this network for later usage you will need to save it to a file.
|
Saving the network file |
Although you do not need to save the network file to train and recall it, it is better to do so. This allows you to compare the results of different training parameters, and prevents loss of work if something goes wrong.
To save the network, select Save As from the Network menu.
Here you are able to choose what directory you want to save your network in, and specify the name of the file.
The convention for the file extension of network files is *.wgt.
The only restriction is that the
path cannot contain a ).
|
Viewing the network |
You can press the View Network File button to view your network, or alternatively select View Network from the Network menu.
Note: You need to associate files of type *.wgt with a text editor in Windows Explorer in order to view the files through the FuzzyCOPE GUI.
Below is an example of a network file.
[FormatVersion = 1.0] /* This is a standard version number. Any value larger than 1.0 will not work*/
[NeuralNetwork]
[NetworkType = Kohonen] /*This shows the type of network being used - MLP / FuNN / Kohonen*/
[NetworkInformation] /*This section of information sets out the layout of the network*/
[NumberOfLayers = 3] /*There are 2 layers (input and map layers) + the bias layer, even though it is not used in Kohonen*/
[InputLayer = 1] /*The input layer is always layer 1*/
[OutputLayer = 2] /*The output layer is the last layer*/
[NumberOfInputs = 4]
[NumberOfOutputs = 9] /*There is one output for each neuron in the map*/
[~NetworkInformation]
[Dimensions = 2] /*This sets out the dimensions of the kohonen map*/
[Dimension=0 Size=3 ]
[Dimension=1 Size=3 ]
[~Dimensions]
[Layer = 0] /*Layer 0 is always the bias layer and is included even though it is not used in Kohonen*/
[LayerType = Simple] /*The type of the bias layer is always simple*/
[NumberOfNeurons = 0]
[All ActivationFunction=error ] /*This is set to error because the bias layer not be activated in Kohonen*/
[~Layer]
[Layer = 1] /*Layer 1 is always the input layer*/
[LayerType = Input]
[NumberOfNeurons = 4]
[All ActivationFunction=linear ]
[~Layer]
[Layer = 2] /*The output layer*/
[LayerType = Simple]
[NumberOfNeurons = 9]
[All ActivationFunction=linear ]
[~Layer]
[ConnectionLayer]
[LayerFrom = 1]
[LayerTo = 2]
[ConnectionType = FullyConnected]
[Attributes UnPrunable UnFreezable ]
[Matrix Weights ]
/*Below is the matrix of connection weights for this connection layer*/
/*Note: the weight values have been rounded to 6dp for formatting purposes*/
| /* To-> 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | From*/ |
| -0.929441 | 0.193945 | 0.298563 | 0.107028 | -0.492660 | 0.132481 | 0.897397 | 0.430158 | -0.664052 | /*0*/ |
| -0.738823 | -0.493576 | 0.772515 | 0.021455 | -0.565966 | -0.734306 | -0.574572 | 0.464827 | -0.335246 | /*1*/ |
| -0.083956 | -0.770806 | 0.751152 | 0.241676 | -0.528001 | 0.228431 | 0.616504 | 0.103916 | 0.739067 | /*2*/ |
| -0.032014 | 0.402692 | -0.914487 | 0.542222 | -0.329020 | -0.937864 | 0.258217 | 0.143223 | 0.648915 | /*3*/ |
[~NeuralNetwork]
|
Opening an existing network |
It is possible to load an existing network by selecting Load Network from the Network menu.
The file can have any extension, but the convention is for network files to have the extension *.wgt.
The only restriction is that the path cannot contain a ).