MLPGA


Function

This app will perform genetic algorithm training on the specified MLP.

Usage

MLPGA parameterfile

Parameter File Example

WeightFile = iris.wgt
DestinationFile = iris-gen.wgt
#DataClasses = 1
DataFile = iris.trn
PopulationSize = 100
Generations = 100
EliteIndividuals = 1
MutationRate = 0.01
CrossoverPoints = 1
Selection = Tournament
Normalisation = on
MaxWeight = 20.0
MinWeight = -20.0

Parameter File Explained

WeightFile : file to load MLP to train from
DestinationFile : file to save trained MLP to
#DataClasses : number of training data classes
DataFile : name of a file to load, must be one of these tags for each data class
PopulationSize : number of individuals in GA population
Generations : number of generations to run GA for
EliteIndividuals : number of elite individuals
MutationRate : rate of mutation
CrossOver : number of points to crossover during reproduction
Selection : selection strategy to use, must be Roulette or Tournament
MaxWeight : maximum weight value allowable
MinWeight : minimum weight value allowable


return to main page


This page is maintained by Michael Watts (http://mike.watts.net.nz)
Last modified on: 12/10/98.