Network Class Reference
[Math]
Backpropagation neural network with optional momentum and adaptive learn rate. More...
#include <NeuralNetwork.h>
Inheritance diagram for Network:
Public Member Functions | |
| Network (int inlen, int len1, int len2, int outlen) | |
| Network (int id, Connection &conn) | |
| Network (const Network &net) | |
| void | Write (Connection &conn) |
| double | CalcError (const Array< double > &input, const Array< double > &output) |
| double | Train (const Array< double > &input, const Array< double > &output, double dErrTarget, int maxCycles) |
| void | Activate (const Array< double > &input, Array< double > &output) |
| int | NetworkId () const |
| int | InputCount () const |
| int | Layer1Count () const |
| int | Layer2Count () const |
| int | OutputCount () const |
| void | ResetID () |
| Network & | operator= (const Network &net) |
| void | ValidateMem () const |
| void | CheckMem () const |
Protected Member Functions | |
| void | Init () |
| void | ActivateLayer (Array< double > &layer1, Array< double > &layer2, Array< double > &wt) |
| layer1 is "behind" layer 2. If layer1 is Input, then layer2 is hidden layer 1. | |
| void | Activate () |
Protected Attributes | |
| int | m_networkId |
| double | m_learnRate |
| double | m_momentum |
| double | m_wtRange |
| bool | m_useAdaptiveLR |
| bool | m_useAdaptiveMom |
| double | m_previousError |
| Array< double > | m_input |
| Array< double > | m_output |
Detailed Description
Backpropagation neural network with optional momentum and adaptive learn rate.
Definition at line 49 of file NeuralNetwork.h.
Constructor & Destructor Documentation
| Network::Network | ( | int | inlen, | |
| int | len1, | |||
| int | len2, | |||
| int | outlen | |||
| ) |
Normal init
Definition at line 142 of file BackPropNeuralNetwork.cpp.
| Network::Network | ( | int | id, | |
| Connection & | conn | |||
| ) |
Datebase load
Definition at line 185 of file BackPropNeuralNetwork.cpp.
Member Function Documentation
| void Network::ActivateLayer | ( | Array< double > & | layer1, | |
| Array< double > & | layer2, | |||
| Array< double > & | wt | |||
| ) | [protected] |
layer1 is "behind" layer 2. If layer1 is Input, then layer2 is hidden layer 1.
Definition at line 344 of file BackPropNeuralNetwork.cpp.
The documentation for this class was generated from the following files:
- spl/math/NeuralNetwork.h
- src/BackPropNeuralNetwork.cpp