IGLib 1.4
The IGLib base library for development of numerical, technical and business applications.

IG::Neural::NeuralTrainingParametersDto Class Reference

Transfer Object (DTO) for neural network training set. More...

Inheritance diagram for IG::Neural::NeuralTrainingParametersDto:
Collaboration diagram for IG::Neural::NeuralTrainingParametersDto:

List of all members.

Public Member Functions

override NeuralTrainingParameters CreateObject ()
 Creates and returns a new object of the corresponding type.

Public Attributes

double LearningRate
 Learning rate.
double Momentum
 Momentum. Specifies how much changes of weight in the previous iterations affect changes in the current iterations.
double SigmoidAlphaValue
 Sigmoid alpha value (used in networks with sigmoid activation functions).
int InputLength
 Gets or sets the number of input neurons.
int OutputLength
 Gets or sets the number of output neurons.
double InputBoundSafetyFactor
double OutputBoundSafetyFactor
int MaxEpochs
 Maximal number of epochs performed in the training procedure.
int EpochsInBundle
 Number of epochs in bundle (i.e. number of epochs performed at once, without any checks or output operations between). This parameter does not affect the training procedure in terms of results.
VectorDtoBase InputRange
 Range from actual inputs.
VectorDtoBase OutputRange
 Range from actual outputs.
VectorDtoBase ToleranceRms
 Tolerance over RMS error of outputs over training points. Training will continue until error becomes below tolerance or until maximal number of epochs is reached. If less or equal than 0 then this tolerance is not taken into account.
VectorDtoBase ToleranceRmsRelativeToRange
 Relative tolerances on RMS errors of outputs over training points, relative to the correspoinding ranges of output data.
double ToleranceRmsRelativeToRangeScalar = NeuralTrainingParameters.DefaultToleranceRmsRelativeToRangeScalar
 Scalar through which all components of the Relative tolerances on RMS errors of outputs can be set to the same value.
VectorDtoBase ToleranceMax
 Tolerance on maximal error of outputs over training points. Training will continue until error becomes below tolerance or until maximal number of epochs is reached. If less or equal than 0 then this tolerance is not taken into account.
VectorDtoBase ToleranceMaxRelativeToRange
 Relative tolerances on max. abs. errors of outputs over training points, relative to the correspoinding ranges of output data.
double ToleranceMaxRelativeToRangeScalar = NeuralTrainingParameters.DefaultToleranceMaxRelativeToRangeScalar
 Scalar through which all components of the Relative tolerances on max. abs. errors of outputs can be set to the same value.
int NumHiddenLayers
 Number of Hideden layers.
int[] NumHidenNeurons
 Numbers of neurons in each hidden layer.
bool IsNetworkTrained
 Whether the network is trained (and results exist).
VectorDtoBase ErrorsTrainingRms
 RMS errors calculated on training data.
VectorDtoBase[] ErrorsTrainingRmsTable
 Convergence Table of RMS errors calculated on training data.
VectorDtoBase ErrorsTrainingMax
 Maximal errors calculated on training data.
VectorDtoBase[] ErrorsTrainingMaxTable
 Convergence Table of Maximal errors calculated on training data.
VectorDtoBase ErrorsTrainingMeanAbs
 Mean absolute errors calculated on training data.
VectorDtoBase ErrorsVerificationRms
 RMS errors calculated on verification data.
VectorDtoBase[] ErrorsVerificationRmsTable
 Convergence Table of RMS errors calculated on verification data.
VectorDtoBase ErrorsVerificationMax
 Maximal errors calculated on verification data.
VectorDtoBase[] ErrorsVerificationMaxTable
 Convergence Table of Maximal errors calculated on verification data.
VectorDtoBase ErrorsVerificationMeanAbs
 Maximal errors calculated on training data.
List< VectorDtoBaseErrorsRmsList
int NumEpochs
 Number of epochs actually spent at training.
double TrainingTime
 Time spent for training.
double TrainingCpuTime
 CPU time spent for training.
int[] EpochNumbers
 List of epoch numbers at which convergence data was sampled.
double[] EpochErrorsRms
 List of sampled RMS errors corresponding to epoch numbers from EpochNumbers.
double[] EpochErrorsAbs
 List of sampled absolute errors corresponding to epoch numbers from EpochNumbers.

Protected Member Functions

override void CopyFromPlain (NeuralTrainingParameters trainingParameters)
 Copies the specified training parameters to the current DTO.
override void CopyToPlain (ref NeuralTrainingParameters trainingParameters)
 Copies contents of the current DTO to the specified training parameters object.

Detailed Description

Transfer Object (DTO) for neural network training set.

$A Igor Jun12; Tako78 Jun12;


Member Function Documentation

override NeuralTrainingParameters IG::Neural::NeuralTrainingParametersDto::CreateObject ( ) [inline]

Creates and returns a new object of the corresponding type.

override void IG::Neural::NeuralTrainingParametersDto::CopyFromPlain ( NeuralTrainingParameters  trainingParameters) [inline, protected]

Copies the specified training parameters to the current DTO.

Parameters:
trainingParametersObject that is copied to the current DTO.
override void IG::Neural::NeuralTrainingParametersDto::CopyToPlain ( ref NeuralTrainingParameters  trainingParameters) [inline, protected]

Copies contents of the current DTO to the specified training parameters object.

Parameters:
trainingParametersObject that the current DTO content is copied to.

Member Data Documentation

Momentum. Specifies how much changes of weight in the previous iterations affect changes in the current iterations.

Sigmoid alpha value (used in networks with sigmoid activation functions).

Gets or sets the number of input neurons.

Gets or sets the number of output neurons.

Maximal number of epochs performed in the training procedure.

$A Tako78 Jul12;

Number of epochs in bundle (i.e. number of epochs performed at once, without any checks or output operations between). This parameter does not affect the training procedure in terms of results.

$A Tako78 Jul12;

Range from actual inputs.

$A Tako78 Octl12;

Range from actual outputs.

$A Tako78 Octl12;

Tolerance over RMS error of outputs over training points. Training will continue until error becomes below tolerance or until maximal number of epochs is reached. If less or equal than 0 then this tolerance is not taken into account.

$A Tako78 Jul12; Igor Jul12;

Relative tolerances on RMS errors of outputs over training points, relative to the correspoinding ranges of output data.

See also:
NeuralApproximatorBase.ToleranceRmsRelativeToRange

Scalar through which all components of the Relative tolerances on RMS errors of outputs can be set to the same value.

See also:
NeuralApproximatorBase.ToleranceRmsRelativeToRangeScalar

Tolerance on maximal error of outputs over training points. Training will continue until error becomes below tolerance or until maximal number of epochs is reached. If less or equal than 0 then this tolerance is not taken into account.

$A Tako78 Jul12;

Relative tolerances on max. abs. errors of outputs over training points, relative to the correspoinding ranges of output data.

See also:
NeuralApproximatorBase.ToleranceMaxRelativeToRange

Scalar through which all components of the Relative tolerances on max. abs. errors of outputs can be set to the same value.

See also:
NeuralApproximatorBase.ToleranceMaxRelativeToRangeScalar

Number of Hideden layers.

$A Tako78 Jul12;

Numbers of neurons in each hidden layer.

$A Tako78 Jul12;

Whether the network is trained (and results exist).

$A Tako78 Jul12;

RMS errors calculated on training data.

$A Tako78 Jul12;

Convergence Table of RMS errors calculated on training data.

$A Tako78 Aug12;

Maximal errors calculated on training data.

$A Tako78 Jul12;

Convergence Table of Maximal errors calculated on training data.

$A Tako78 Aug12;

Mean absolute errors calculated on training data.

$A Tako78 Jul12;

RMS errors calculated on verification data.

$A Tako78 Jul12;

Convergence Table of RMS errors calculated on verification data.

$A Tako78 Aug12;

Maximal errors calculated on verification data.

$A Tako78 Jul12;

Convergence Table of Maximal errors calculated on verification data.

$A Tako78 Aug12;

Maximal errors calculated on training data.

$A Tako78 Jul12;

Number of epochs actually spent at training.

$A Tako78 Jul12;

Time spent for training.

$A Tako78 Jul12;;

CPU time spent for training.

$A Tako78 Jul12;

List of epoch numbers at which convergence data was sampled.

List of sampled RMS errors corresponding to epoch numbers from EpochNumbers.

List of sampled absolute errors corresponding to epoch numbers from EpochNumbers.


The documentation for this class was generated from the following file:
 All Classes Namespaces Files Functions Variables Enumerations Properties Events