Neural Networks in R using the Stuttgart Neural Network Simulator (SNNS)


[Up] [Top]

Documentation for package ‘RSNNS’ version 0.3-1

Help Pages

A C D E G I J M N O P R S T V W

-- --

RSNNS-package Getting started with the RSNNS package

-- --

$ Enable calling of C++ functions as methods of SnnsR-class objects.
$-method Enable calling of C++ functions as methods of SnnsR-class objects.

-- A --

analyzeClassification Converts continuous outputs to class labels.
art1 Create and train an art1 network.
art1.default Create and train an art1 network.
art2 Create and train an art2 network.
art2.default Create and train an art2 network.
assoz Create and train an (auto-)associative memory.
assoz.default Create and train an (auto-)associative memory.

-- C --

confusionMatrix Computes a confusion matrix.
createNet$getAllUnitsTType Create a layered network.
createNet-method Create a layered network.
createPatSet-method Create a pattern set.

-- D --

decodeClassLabels Decode class labels from a numerical or levels vector to a binary matrix.
dlvq Create and train a dlvq network.
dlvq.default Create and train a dlvq network.

-- E --

elman Create and train an elman network.
elman.default Create and train an elman network.
encodeClassLabels Applies analyzeClassification row-wise to a matrix.
extractPatterns-method Extract the current pattern set to a matrix.

-- G --

genericPredictCurrPatSet-method Predict values with a trained net.
getAllHiddenUnits-method Get all hidden units of the net.
getAllInputUnits-method Get all input units of the net.
getAllOutputUnits-method Get all output units of the net.
getAllUnitsTType-method Get all units in the net of a certain ttype.
getSnnsRDefine Get a define of the SNNS kernel.
getSnnsRFunctionTable Get the function table of available SNNS functions.
getUnitsByName-method Find all units whose name begins with a given prefix.

-- I --

initializeNet-method Initialize the network.
inputColumns Function to get the columns that are inputs.

-- J --

jordan Create and train a jordan network.
jordan.default Create and train a jordan network.

-- M --

matrixToActMapList Organize a matrix containing 1d vectors of network activations as 2d maps.
mlp Create and train a multi-layer perceptron (mlp).
mlp.default Create and train a multi-layer perceptron (mlp).

-- N --

normalizeData Data normalization.

-- O --

outputColumns Function to get the columns that are targets.

-- P --

plotActMap Plot an activation map as a heatmap.
plotIterativeError Plot the iterative training and test error of the net of this rsnns object.
plotIterativeError.reg_class Plot the iterative training and test error of the net of this rsnns object.
plotRegressionError Plot a regression error plot.
plotROC Plot a ROC curve.
predict.rsnns Generic predict function for rsnns object.
predictCurrPatSet-method Predict values with a trained net.
print.rsnns Generic print function for rsnns objects.

-- R --

rbf Create and train a radial basis function (rbf) network.
rbf.default Create and train a radial basis function (rbf) network.
rbfDDA Create and train a rbf network with the DDA algorithm.
rbfDDA.default Create and train a rbf network with the DDA algorithm.
readPatFile Load data from a pat file.
readResFile Rudimentary parser for res files.
resetRSNNS-method Reset the SnnsR-class object.
resolveSnnsRDefine Resolve a define of the SNNS kernel.
RSNNS Getting started with the RSNNS package
rsnns Object factory for generating rsnns objects.
rsnnsObjectFactory Object factory for generating rsnns objects.

-- S --

savePatFile Save data to a pat file.
setSnnsRSeedValue Set the seed value used in all SnnsR objects.
setTTypeUnitsActFunc-method Set the activation function for all units of a certain ttype.
setUnitDefaults-method Set the unit defaults.
snnsData Example data of the package.
SnnsR The main class of the package.
SnnsR-class The main class of the package.
SnnsRObject$createPatSet Create a pattern set.
SnnsRObject$extractPatterns Extract the current pattern set to a matrix.
SnnsRObject$genericPredictCurrPatSet Predict values with a trained net.
SnnsRObject$getAllHiddenUnits Get all hidden units of the net.
SnnsRObject$getAllInputUnits Get all input units of the net.
SnnsRObject$getAllOutputUnits Get all output units of the net.
SnnsRObject$getAllUnitsTType Get all units in the net of a certain ttype.
SnnsRObject$getUnitsByName Find all units whose name begins with a given prefix.
SnnsRObject$initializeNet Initialize the network.
SnnsRObject$predictCurrPatSet Predict values with a trained net.
SnnsRObject$resetRSNNS Reset the SnnsR-class object.
SnnsRObject$setTTypeUnitsActFunc Set the activation function for all units of a certain ttype.
SnnsRObject$setUnitDefaults Set the unit defaults.
SnnsRObject$somPredictComponentMaps Calculate the som component maps.
SnnsRObject$somPredictCurrPatSetWinners Get most of the relevant results from a som.
SnnsRObject$somPredictCurrPatSetWinnersSpanTree Get the spanning tree of the som, calculated directly by SNNS.
SnnsRObject$train Train a network and test it in every training iteration.
SnnsRObject$whereAreResults Get a list of output units of a net.
SnnsRObjectFactory Object factory to create a new object of type SnnsR-class.
SnnsR__createNet Create a layered network.
SnnsR__createPatSet Create a pattern set.
SnnsR__extractPatterns Extract the current pattern set to a matrix.
SnnsR__genericPredictCurrPatSet Predict values with a trained net.
SnnsR__getAllHiddenUnits Get all hidden units of the net.
SnnsR__getAllInputUnits Get all input units of the net.
SnnsR__getAllOutputUnits Get all output units of the net.
SnnsR__getAllUnitsTType Get all units in the net of a certain ttype.
SnnsR__getUnitsByName Find all units whose name begins with a given prefix.
SnnsR__initializeNet Initialize the network.
SnnsR__predictCurrPatSet Predict values with a trained net.
SnnsR__resetRSNNS Reset the SnnsR-class object.
SnnsR__setTTypeUnitsActFunc Set the activation function for all units of a certain ttype.
SnnsR__setUnitDefaults Set the unit defaults.
SnnsR__somPredictComponentMaps Calculate the som component maps.
SnnsR__somPredictCurrPatSetWinners Get most of the relevant results from a som.
SnnsR__somPredictCurrPatSetWinnersSpanTree Get the spanning tree of the som, calculated directly by SNNS.
SnnsR__train Train a network and test it in every training iteration.
SnnsR__whereAreResults Get a list of output units of a net.
som Create and train a self-organizing map (som).
som.default Create and train a self-organizing map (som).
somPredictComponentMaps-method Calculate the som component maps.
somPredictCurrPatSetWinners-method Get most of the relevant results from a som.
somPredictCurrPatSetWinnersSpanTree-method Get the spanning tree of the som, calculated directly by SNNS.
splitForTrainingAndTest Function to split data into training and test set.
summary.rsnns Generic summary function for rsnns objects.

-- T --

toNumericClassLabels Converts a vector (of class labels) to a numeric vector.
train Generic train function.
train-method Train a network and test it in every training iteration.
train.rsnns Generic train function.

-- V --

vectorToActMap Organize network activation as 2d map.

-- W --

whereAreResults-method Get a list of output units of a net.