Defines how a neighborhood function should work in competitive training. This is most often used in the training process for a self-organizing map. This function determines to what degree the training should take place on a neuron, based on its proximity to the "winning" neuron.
For a list of all members of this type, see INeighborhoodFunction Members .
| Type | Description |
|---|---|
| NeighborhoodBubble | A neighborhood function that uses a simple bubble. A width is defined, and any neuron that is plus or minus that width from the winning neuron will be updated as a result of training. |
| NeighborhoodGaussian | A neighborhood function based on the Gaussian function. |
| NeighborhoodSingle | A very simple neighborhood function that will return 1.0 (full effect) for the winning neuron, and 0.0 (no change) for everything else. |
Namespace: Encog.Neural.Networks.Training.Competitive.Neighborhood
Assembly: encog-core-cs (in encog-core-cs.dll)
INeighborhoodFunction Members | Encog.Neural.Networks.Training.Competitive.Neighborhood Namespace