/mlp_backpropagation

Multi layer perceptron backpropagation

Primary LanguageJava

Multi Layer Perceptron Backpropagation (1 hidden layer) example

Based on the great article: Neural network backpropagation with Java

Other resources: A Step by Step Backpropagation Example

PARAMETERS

FUNCTION APPROXIMATION ARGUMENTS

in=1 inf=1 hidden=20 out=1 outf=2 maxEpoch=200000 printApproximation=true printError=false trainDSPath=data\approximation_train_1.txt testDSPath=data\approximation_test.txt

TRANSFORMATION

in=4 inf=1,2,3,4 hidden=2 out=4 outf=1,2,3,4 maxEpoch=1500000 normalizeFeatures=false printApproximation=false printError=true trainDSPath=data\transformation.txt testDSPath=data\transformation.txt

CLASSIFICATION

in=4 inf=1,2,3,4 hidden=8 out=3 outf=5 outFeatureAs01=true maxEpoch=50000 normalizeFeatures=false printApproximation=false printError=true trainDSPath=data\classification_train.txt testDSPath=data\classification_test.txt