Network

Network
 
The Network view shows how the nodes in an EasyNN-plus neural network are interconnected.  When this view is opened the Network Editor starts.  Any source node can be selected with the left mouse button and any destination node can be selected with right mouse button.  The network nodes and connections can be edited to produce any configuration.
 
How to create a new neural network
 
A new neural network can be created from the Grid by pressing the New Network toolbar button or selecting Action > New Network.  This will produce the New Network dialog.  This dialog allows the neural network configuration to be specified.  The dialog will already contain the necessary information to generate a neural network that will be capable of learning the information in the Grid.  However, the generated network may take a long time to learn and it may give poor results when tested.  A better neural network can be generated by checking Grow hidden layer 1 and allowing EasyNN-plus to determine the optimum number of nodes and connections. 
 
It is rarely necessary to have more than one layer of hidden nodes but EasyNN-plus will generate two or three hidden layers if Grow hidden layer 2 and Grow hidden layer 3 are checked. 
 
The time that EasyNN-plus will spend looking for the optimum network can be controlled by setting the Growth rate variables.  Every time that the period expires EasyNN-plus will generate a new neural network slightly different from the previous one.  The best network is saved.
 
How to use the Network editor
 
To create a network manually using the Network Editor start with a suitable Grid and then use New Network but do not check Connect layers.  This will create the optimum number of nodes but the weights between the layers will not be connected.
 
To connect and disconnect nodes the source and the destination of the weight connections need to be selected.  This is done using the mouse.  The left button selects the source and the right button selects the destination.  First left click on a node to select the source and then right click on a node to select the destination.  The selected source node will now have a wide red border.  The other nodes in the source layer will have a narrow red border.  The destination node will now have a wide blue border.  The other nodes in the destination layer will have a narrow blue border.  The right click will also open a menu of functions that are used to change the network connections or add and delete nodes.
 
Any nodes or layers can be connected to any other nodes or layers.  Feed forward, feedback and skip layer connections are possible.
 
The number of hidden layer nodes can be increased or decreased and hidden layers can be added if needed.  The nodes will be reconnected using the connections that are held in the slave memory.  These connections will be as close as possible to its previous state.  It can then be edited further as required.
 
1

Input Node

1. Input NodeInput nodes are connected to the input columns in the grid.
 
2

Hidden Node

2. Hidden NodeHidden nodes are fully connected to input nodes, output nodes or other layers of hidden nodes.
 
3

Output Node

3. Output NodeOutput nodes are connected to the output columns in the grid.
 
4

Connections

4. ConnectionsThe connections between the nodes are weighted.  The color represents the polarity and the thickness represents the weight value.  The weight values and polarity change as the network learns.
 
5

Network Edit Menu

5. Network Edit Menu
 
Left click menu item required. 

Created with help of DrExplain