SuperNN  1.0.0
Public Member Functions | List of all members
SuperNN::Incremental Struct Reference

Incremental backpropagation. More...

Inheritance diagram for SuperNN::Incremental:
Inheritance graph
[legend]

Public Member Functions

 Incremental ()
 
virtual ~Incremental ()
 
virtual unsigned train (Network &net, const Data &data, double dmse=0, unsigned max_epochs=1000)
 Adjusts the synaptic weight of an artificial neural network in order to minimize the error (MSE for standard l2 training or MAE for specific l1 training). More...
 
- Public Member Functions inherited from SuperNN::ImplBackprop
 ImplBackprop ()
 
virtual ~ImplBackprop ()
 
- Public Member Functions inherited from SuperNN::TrainingAlgorithm
virtual ~TrainingAlgorithm ()
 
virtual void clear_derror_acc (Network &net)
 Clears the accumulated error partial derivatives. More...
 
virtual double delta (Network &net, unsigned l, unsigned n)
 Calculates the local error gradient for each neuron. More...
 
virtual void derror_acc (Network &net)
 Accumulates the error partial derivative in respect to the weights, for each connection of the neural network. More...
 
virtual void prepare (Network &net)
 Prepares the trainer and a neural network for training. More...
 

Additional Inherited Members

- Public Attributes inherited from SuperNN::ImplBackprop
double eta
 Initial learning rate. More...
 
double eta_df
 Learning rate decrease factor (must be <= 1) More...
 
double eta_if
 Learning rate increase factor (must be >= 1) More...
 
double eta_max
 Maximum learning rate. More...
 
double eta_min
 Minimum learning rate. More...
 
- Protected Member Functions inherited from SuperNN::ImplBackprop
virtual void update_eta (double mse, double last_mse)
 Calculates the new learning rate (and updates it), based on the mean squared error last change. More...
 
virtual void update_weights (Network &net, double factor=1)
 Updates the weights using the accumulated error partial derivative calculated by derror_acc(). More...
 
- Protected Member Functions inherited from SuperNN::TrainingAlgorithm
virtual void check (const Network &net, const Data &data) const
 Checks if the dimensions match and if the training algorithm can be used with a given network and data. More...
 

Detailed Description

Incremental backpropagation.

The weights are changed after each presented training pattern.

Definition at line 160 of file training.hpp.

Constructor & Destructor Documentation

SuperNN::Incremental::Incremental ( )

Definition at line 161 of file training.cpp.

SuperNN::Incremental::~Incremental ( )
virtual

Definition at line 165 of file training.cpp.

Member Function Documentation

unsigned SuperNN::Incremental::train ( Network net,
const Data data,
double  dmse = 0,
unsigned  max_epochs = 1000 
)
virtual

Adjusts the synaptic weight of an artificial neural network in order to minimize the error (MSE for standard l2 training or MAE for specific l1 training).

Parameters
netArtificial neural network to be trained
dataTraining data
dmseDesired mean squared error (or MAE when applied). Stopping condition
max_epochsMaximum number of epochs to train the network. Stopping condition
Returns
The number of training epochs performed

Implements SuperNN::ImplBackprop.

Definition at line 169 of file training.cpp.