Documentation For Encog 2.x

Backpropagation Members

Backpropagation overview

Public Instance Constructors

Backpropagation Overloaded. Initializes a new instance of the Backpropagation class.

Public Instance Properties

Error (inherited from BasicTraining) Get the current error percent from the training.
LearningRate The learning rate, this is value is essentially a percent. It is the degree to which the gradients are applied to the weight matrix to allow learning.
Levels (inherited from Propagation) The propagation levels.
Momentum The momentum for training. This is the degree to which changes from which the previous training iteration will affect this training iteration. This can be useful to overcome local minima.
Network (inherited from Propagation) Get the current best neural network.
OutputHolder (inherited from Propagation) The output holder being used.
Strategies (inherited from BasicTraining) The strategies to use.
Training (inherited from BasicTraining) The training data to use.

Public Instance Methods

AddStrategy (inherited from BasicTraining) Training strategies can be added to improve the training results. There are a number to choose from, and several can be used at once.
BackwardPass (inherited from Propagation) Calculate the error for the recognition just done.
Equals (inherited from Object) 
GetHashCode (inherited from Object) 
GetType (inherited from Object) 
Iteration (inherited from Propagation) Perform one iteration of training. Note: if you get a StackOverflowError while training, then you have endless recurrent loops. Try inserting no trainable synapses on one side of the loop.
PostIteration (inherited from BasicTraining) Call the strategies after an iteration.
PreIteration (inherited from BasicTraining) Call the strategies before an iteration.
ToString (inherited from Object) 

Protected Instance Methods

Finalize (inherited from Object) 
MemberwiseClone (inherited from Object) 

See Also

Backpropagation Class | Encog.Neural.Networks.Training.Propagation.Back Namespace