RestrictedBolzmannMachine: Initialize a Restricted Bolzman Machine

In xrobin/DeepLearning: Deep Learning of neural networks

Description

Creates a Restricted Bolzman Machine (RBM), precisely a RestrictedBolzmannMachine object, with the given specifications. It is typically stacked in a DeepBeliefNet.

Usage

RestrictedBolzmannMachine(input, output, weights = NULL,
  initialize = c("0", "uniform"))

Arguments

input, output	Layer objects
weights	optional starting weights. If NULL, weights will be initialized to 0
initialize	whether to initialize weights and biases with 0 or random uniform values. Ignored if weights are provided.

Value

an object of class RestrictedBolzmannMachine containing the following elements:

• input,outputLayers

• weights, weights.breaksas input

• pretrainedboolean switch to mark the state of the layer. FALSE just after initialization.

Copying/Cloning

For performance purposes, the weights are stored in an environment. This means that when you copy the RestrictedBolzmannMachine with an assignment, you do not copy the weights and any modification you make to the new object will be propagated to the original one, and reciprocally. Use clone to control this and make a copy of the weights whenever you need it. Note that all the functions defined in the package do this by default.

See Also

DeepBeliefNet, which makes use of RestrictedBolzmannMachine objects, Layer.

Examples

rbm <- RestrictedBolzmannMachine(Layer(784, "continuous"), Layer(1000, "binary"))
print(rbm)
methods(class="RestrictedBolzmannMachine")

xrobin/DeepLearning documentation built on Sept. 18, 2020, 5:23 a.m.