Package for Deep Architectures and Restricted Boltzmann Machines

Documentation for package ‘darch’ version 0.10.0

DESCRIPTION file.

Help Pages

A B C D F G I L M N P Q R S T V

-- A --

applyDropoutMask

Applies the given dropout mask to the given data row-wise.

-- B --

backpropagation	Backpropagation learning function
binSigmoidUnit	Binary sigmoid unit function.

-- C --

createDataSet	Create data set using data, targets, a formula, and possibly an existing data set.
createDataSet-method	Create new 'DataSet' by filling an existing one with new data.
createDataSet-method	Create 'DataSet' using data and targets.
createDataSet-method	Constructor function for 'DataSet' objects.
createDataSet.DataSet	Create new 'DataSet' by filling an existing one with new data.
createDataSet.default	Create 'DataSet' using data and targets.
createDataSet.formula	Constructor function for 'DataSet' objects.
crossEntropyError	Cross entropy error function

-- D --

DArch	Class for deep architectures
darch	Fit a deep neural network.
DArch-class	Class for deep architectures
darch.DataSet	Create and train DArch object using a 'DataSet'.
darch.default	Fit deep neural network.
darch.formula	Fit a deep neural network using a formula and a single data frame or matrix.
DataSet	Class for specifying datasets.
dataset	Class for specifying datasets.
DataSet-class	Class for specifying datasets.

-- F --

fineTuneDArch	Fine tuning function for the deep architecture
fineTuneDArch-method	Fine tuning function for the deep architecture

-- G --

generateDropoutMask	Dropout mask generator function.
generateRBMs	Generates the RBMs for the pre-training.
generateRBMs-method	Generates the RBMs for the pre-training.
generateWeights	Generates a weight matrix.
getBatchSize	Returns the batch size of the 'Net'.
getBatchSize-method	Returns the batch size of the 'Net'.
getCancel	Returns the cancel value
getCancel-method	Returns the cancel value
getCancelMessage	Returns the cancel message
getCancelMessage-method	Returns the cancel message
getDropoutHiddenLayers	Returns the dropout rate for the hidden layers
getDropoutHiddenLayers-method	Returns the dropout rate for the hidden layers
getDropoutInputLayer	Returns the dropout rate for the input layer
getDropoutInputLayer-method	Returns the dropout rate for the input layer
getDropoutMask	Returns the dropout mask for the given layer
getDropoutMask-method	Returns the dropout mask for the given layer
getDropoutMasks	Returns the dropout masks
getDropoutMasks-method	Returns the dropout masks
getDropoutOneMaskPerEpoch	Return the dropout usage
getDropoutOneMaskPerEpoch-method	Return the dropout usage
getEpochs	Returns the number of epochs the 'Net' was trained for
getEpochs-method	Returns the number of epochs the 'Net' was trained for
getErrorFunction	Returns the error function of the 'Net'.
getErrorFunction-method	Returns the error function of the 'Net'.
getExecOutput	Returns the execution output of the layer from the 'DArch' object
getExecOutput-method	Returns the execution output of the layer from the 'DArch' object
getExecOutputs	Returns the execution output list of the 'DArch' object
getExecOutputs-method	Returns the execution output list of the 'DArch' object
getExecuteFunction	Returns the function for the execution of the 'DArch' network
getExecuteFunction-method	Returns the function for the execution of the 'DArch' network
getFF	Returns if the weights are saved as ff objects
getFF-method	Returns if the weights are saved as ff objects
getFinalMomentum	Returns the final momentum of the 'Net'.
getFinalMomentum-method	Returns the final momentum of the 'Net'.
getFineTuneFunction	Returns the fine tune function for the 'DArch' object
getFineTuneFunction-method	Returns the fine tune function for the 'DArch' object
getGenWeightFunction	Returns the function for generating weight matrices.
getGenWeightFunction-method	Returns the function for generating weight matrices.
getHiddenBiases	Returns the biases of the hidden units.
getHiddenBiases-method	Returns the biases of the hidden units.
getHiddenBiasesInc	Returns the update value for the biases of the hidden units.
getHiddenBiasesInc-method	Returns the update value for the biases of the hidden units.
getHiddenUnitStates	Returns a list with the states of the hidden units.
getHiddenUnitStates-method	Returns a list with the states of the hidden units.
getInitialMomentum	Returns the momentum of the 'Net'
getInitialMomentum-method	Returns the momentum of the 'Net'
getLayer	Returns a layer from the 'DArch' object
getLayer-method	Returns a layer from the 'DArch' object
getLayerField	Returns the field of a layer from the 'DArch' object
getLayerField-method	Returns the field of a layer from the 'DArch' object
getLayerFunction	Returns the neuron function of a layer from the 'DArch' object
getLayerFunction-method	Returns the neuron function of a layer from the 'DArch' object
getLayers	Returns the a list of layers from the 'DArch' object
getLayers-method	Returns the a list of layers from the 'DArch' object.
getLayerWeights	Returns the weights of a layer from the 'DArch' object
getLayerWeights-method	Returns the weights of a layer from the 'DArch' object
getLearnRateBiases	Returns the learning rate for the bias weights of the 'DArch' object
getLearnRateBiases-method	Returns the learning rate for the bias weights of the 'DArch' object
getLearnRateBiasHidden	Returns the learning rate for the hidden biases.
getLearnRateBiasHidden-method	Returns the learning rate for the hidden biases.
getLearnRateBiasVisible	Returns the learning rate for the visible biases.
getLearnRateBiasVisible-method	Returns the learning rate for the visible biases.
getLearnRateWeights	Returns the learn rate of the weights.
getLearnRateWeights-method	Returns the learn rate of the weights.
getMomentum	Returns the current momentum of the 'Net'.
getMomentum-method	Returns the current momentum of the 'Net'.
getMomentumSwitch	Returns the momentum switch of the 'Net'.
getMomentumSwitch-method	Returns the momentum switch of the 'Net'.
getNormalizeWeights	Returns whether weight normalization is active
getNormalizeWeights-method	Returns whether weight normalization is active
getNumHidden	Returns the number of hidden units of the 'RBM'
getNumHidden-method	Returns the number of hidden units of the 'RBM'
getNumVisible	Returns the number of visible units of the 'RBM'
getNumVisible-method	Returns the number of visible units of the 'RBM'
getOutput	Returns the output of the 'RBM'
getOutput-method	Returns the output of the 'RBM'
getPosPhaseData	Returns the data for the positive phase.
getPosPhaseData-method	Returns the data for the positive phase.
getRBMList	Returns a list of 'RBM's of the 'DArch' object
getRBMList-method	Returns a list of 'RBM's of the 'DArch' object
getStats	Returns the list of statistics for the network
getStats-method	Returns the list of statistics for the network
getVisibleBiases	Returns the biases of the visible units.
getVisibleBiases-method	Returns the biases of the visible units.
getVisibleBiasesInc	Returns the update value for the biases of the visible units.
getVisibleBiasesInc-method	Returns the update value for the biases of the visible units.
getVisibleUnitStates	Returns a list with the states of the visible units.
getVisibleUnitStates-method	Returns a list with the states of the visible units.
getWeightCost	Returns the weigth cost for the training
getWeightCost-method	Returns the weigth cost for the training
getWeightInc	Returns the update value for the weights.
getWeightInc-method	Returns the update value for the weights.
getWeights	Returns the weights of the 'RBM'.
getWeights-method	Returns the weights of the 'RBM'.

-- I --

incrementEpochs	Increment the number of epochs this 'Net' has been trained for
incrementEpochs-method	Increment the number of epochs this 'Net' has been trained for

-- L --

linearUnit	Linear unit function.
linearUnitDerivative	Linear unit function with unit derivatives.
linearUnitFunc	Calculates the linear neuron output no transfer function
loadDArch	Loads a DArch network
loadRBM	Loads a RBM network
loadRBMFFWeights	Loads weights and biases for a RBM network from a ffData file.

-- M --

makeStartEndPoints	Makes start- and end-points for the batches.
maxoutUnitDerivative	Maxout unit function with unit derivatives.
minimize	Minimize a differentiable multivariate function.
minimizeAutoencoder	Conjugate gradient for a autoencoder network
minimizeClassifier	Conjugate gradient for a classification network
mseError	Mean squared error function

-- N --

Net	Abstract class for neural networks.
Net-class	Abstract class for neural networks.
newDArch	Constructor function for 'DArch' objects.
newRBM	Constructor function for RBM object.

-- P --

predict.DArch	Forward-propagate data.
predict.darch	Forward-propagate data.
preTrainDArch	Pre-trains a 'DArch' network
preTrainDArch-method	Pre-trains a 'DArch' network
print.DArch	Print DArch details.
print.darch	Print DArch details.
provideMNIST	Provides MNIST data set in the given folder.

-- Q --

quadraticError

Quadratic error function

-- R --

RBM	Class for restricted Boltzmann machines
RBM-class	Class for restricted Boltzmann machines
rbmUpdate	Function for updating the weights and biases of an 'RBM'
readMNIST	Function for generating ff files of the MNIST Database
removeLayerField	Removes a layer from the 'DArch' object
removeLayerField-method	Removes a layer from the 'DArch' object
resetDArch	Resets the weights and biases of the 'DArch' object
resetDArch-method	Resets the weights and biases of the 'DArch' object
resetExecOutput	Resets the output list of the 'DArch' object
resetExecOutput-method	Resets the output list of the 'DArch' object
resetRBM	Resets the weights and biases of the 'RBM' object
resetRBM-method	Resets the weights and biases of the 'RBM' object
rpropagation	Resilient backpropagation training for deep architectures.
runDArch	Execute the darch

-- S --

saveDArch	Saves a DArch network
saveDArch-method	Saves a DArch network
saveRBM	Saves a RBM network
saveRBM-method	Saves a RBM network
saveRBMFFWeights	Saves weights and biases of a RBM network into a ffData file.
saveRBMFFWeights-method	Saves weights and biases of a RBM network into a ffData file.
setBatchSize	Sets the batch size of the 'Net'.
setBatchSize<-	Sets the batch size of the 'Net'.
setBatchSize<--method	Sets the batch size of the 'Net'.
setCancel	Set whether the learning shall be canceled.
setCancel<-	Set whether the learning shall be canceled.
setCancel<--method	Set whether the learning shall be canceled.
setCancelMessage	Sets the cancel message.
setCancelMessage<-	Sets the cancel message.
setCancelMessage<--method	Sets the cancel message.
setDropoutHiddenLayers	Sets the dropout rate for the hidden layers.
setDropoutHiddenLayers<-	Sets the dropout rate for the hidden layers.
setDropoutHiddenLayers<--method	Sets the dropout rate for the hidden layers.
setDropoutInputLayer	Sets the dropout rate for the input layer.
setDropoutInputLayer<-	Sets the dropout rate for the input layer.
setDropoutInputLayer<--method	Sets the dropout rate for the input layer.
setDropoutMask	Set the dropout mask for the given layer.
setDropoutMask<-	Set the dropout mask for the given layer.
setDropoutMask<--method	Set the dropout mask for the given layer.
setDropoutMasks	Set the dropout masks.
setDropoutMasks<-	Set the dropout masks.
setDropoutMasks<--method	Set the dropout masks.
setDropoutOneMaskPerEpoch<-	Set dropout mask usage
setDropoutOneMaskPerEpoch<--method	Set dropout mask usage
setErrorFunction	Sets the error function of the 'Net'.
setErrorFunction<-	Sets the error function of the 'Net'.
setErrorFunction<--method	Sets the error function of the 'Net'.
setExecuteFunction	Sets the execution function for the network
setExecuteFunction<-	Sets the execution function for the network
setExecuteFunction<--method	Sets the execution function for the network
setFF	Sets if the weights are saved as ff objects
setFF<-	Sets if the weights are saved as ff objects
setFF<--method	Sets if the weights are saved as ff objects
setFinalMomentum	Sets the final momentum of the 'Net'.
setFinalMomentum<-	Sets the final momentum of the 'Net'.
setFinalMomentum<--method	Sets the final momentum of the 'Net'.
setFineTuneFunction	Sets the fine tuning function for the network
setFineTuneFunction<-	Sets the fine tuning function for the network
setFineTuneFunction<--method	Sets the fine tuning function for the network
setGenWeightFunction	Sets the function for generating weight matrices.
setGenWeightFunction<-	Sets the function for generating weight matrices.
setGenWeightFunction<--method	Sets the function for generating weight matrices.
setHiddenBiases	Sets the biases of the hidden units for the 'RBM' object
setHiddenBiases<-	Sets the biases of the hidden units for the 'RBM' object
setHiddenBiases<--method	Sets the biases of the hidden units for the 'RBM' object
setHiddenBiasesInc	Sets the update value for the biases of the hidden units
setHiddenBiasesInc<-	Sets the update value for the biases of the hidden units
setHiddenBiasesInc<--method	Sets the update value for the biases of the hidden units
setHiddenUnitFunction	Sets the unit function of the hidden units
setHiddenUnitFunction<-	Sets the unit function of the hidden units
setHiddenUnitFunction<--method	Sets the unit function of the hidden units
setHiddenUnitStates	Sets the states of the hidden units
setHiddenUnitStates<-	Sets the states of the hidden units
setHiddenUnitStates<--method	Sets the states of the hidden units
setInitialMomentum<-	Sets the initial momentum of the 'Net'
setInitialMomentum<--method	Sets the initial momentum of the 'Net'
setLayer	Sets a layer with the given index for the network
setLayer<-	Sets a layer with the given index for the network
setLayer<--method	Sets a layer with the given index for the network
setLayerField	Sets a field in a layer.
setLayerField<-	Sets a field in a layer.
setLayerField<--method	Sets a field in a layer.
setLayerFunction	Sets the function for a layer with the given index
setLayerFunction<-	Sets the function for a layer with the given index
setLayerFunction<--method	Sets the function for a layer with the given index
setLayers	Sets the layers for the network
setLayers<-	Sets the layers for the network
setLayers<--method	Sets the layers for the network
setLayerWeights	Sets the weights of a layer with the given index
setLayerWeights<-	Sets the weights of a layer with the given index
setLayerWeights<--method	Sets the weights of a layer with the given index
setLearnRateBiases	Sets the learning rate for the biases
setLearnRateBiases<-	Sets the learning rate for the biases
setLearnRateBiases<--method	Sets the learning rate for the biases
setLearnRateBiasHidden	Sets the learning rates of the biases for the hidden units
setLearnRateBiasHidden<-	Sets the learning rates of the biases for the hidden units
setLearnRateBiasHidden<--method	Sets the learning rates of the biases for the hidden units
setLearnRateBiasVisible	Sets the learnig rates of the biases for the visible units
setLearnRateBiasVisible<-	Sets the learnig rates of the biases for the visible units
setLearnRateBiasVisible<--method	Sets the learnig rates of the biases for the visible units
setLearnRateWeights	Sets the learning rate for the weights.
setLearnRateWeights<-	Sets the learning rate for the weights.
setLearnRateWeights<--method	Sets the learning rate for the weights.
setLogLevel	Sets the log level for the 'Net'.
setLogLevel<-	Sets the log level for the 'Net'.
setLogLevel<--method	Sets the log level for the 'Net'.
setMomentumSwitch	Sets the momentum switch of the 'Net'.
setMomentumSwitch<-	Sets the momentum switch of the 'Net'.
setMomentumSwitch<--method	Sets the momentum switch of the 'Net'.
setNormalizeWeights<-	Set whether weight normalization should be performed
setNormalizeWeights<--method	Set whether weight normalization should be performed
setNumHidden	Sets the number of hidden units
setNumHidden<-	Sets the number of hidden units
setNumHidden<--method	Sets the number of hidden units
setNumVisible	Sets the number of visible units
setNumVisible<-	Sets the number of visible units
setNumVisible<--method	Sets the number of visible units
setOutput	Sets the output of the 'RBM' object
setOutput<-	Sets the output of the 'RBM' object
setOutput<--method	Sets the output of the 'RBM' object
setPosPhaseData	Sets the positive phase data for the training
setPosPhaseData<-	Sets the positive phase data for the training
setPosPhaseData<--method	Sets the positive phase data for the training
setRBMList<-	Sets the list of RBMs
setRBMList<--method	Sets the list of RBMs
setStats	Adds a list of statistics to the network
setStats<-	Adds a list of statistics to the network
setStats<--method	Adds a list of statistics to the network
setUpdateFunction	Sets the update function of the 'RBM' object
setUpdateFunction<-	Sets the update function of the 'RBM' object
setUpdateFunction<--method	Sets the update function of the 'RBM' object
setVisibleBiases	Sets the biases of the visible units for the 'RBM' object
setVisibleBiases<-	Sets the biases of the visible units for the 'RBM' object
setVisibleBiases<--method	Sets the biases of the visible units for the 'RBM' object
setVisibleBiasesInc	Sets the update value for the biases of the visible units
setVisibleBiasesInc<-	Sets the update value for the biases of the visible units
setVisibleBiasesInc<--method	Sets the update value for the biases of the visible units
setVisibleUnitFunction	Sets the unit function of the visible units
setVisibleUnitFunction<-	Sets the unit function of the visible units
setVisibleUnitFunction<--method	Sets the unit function of the visible units
setVisibleUnitStates	Sets the states of the visible units
setVisibleUnitStates<-	Sets the states of the visible units
setVisibleUnitStates<--method	Sets the states of the visible units
setWeightCost	Sets the weight costs for the training
setWeightCost<-	Sets the weight costs for the training
setWeightCost<--method	Sets the weight costs for the training
setWeightInc	Sets the update values for the weights
setWeightInc<-	Sets the update values for the weights
setWeightInc<--method	Sets the update values for the weights
setWeights	Sets the weights of the 'RBM' object
setWeights<-	Sets the weights of the 'RBM' object
setWeights<--method	Sets the weights of the 'RBM' object
sigmoidUnit	Sigmoid unit function.
sigmoidUnitDerivative	Sigmoid unit function with unit derivatives.
sigmUnitFunc	Calculates the neuron output with the sigmoid function
sigmUnitFuncSwitch	Calculates the neuron output with the sigmoid function
softmaxUnit	Softmax unit function.
softmaxUnitDerivative	Softmax unit function with unit derivatives.

-- T --

tanSigmoidUnit	Continuous Tan-Sigmoid unit function.
tanSigmoidUnitDerivative	Continuous Tan-Sigmoid unit function.
trainRBM	Trains a 'RBM' with contrastive divergence
trainRBM-method	Trains a 'RBM' with contrastive divergence

-- V --

validateDataSet	Validate 'DataSet'
validateDataSet-method	Validate 'DataSet'