reconstruct: Reconstruct data through a Deep Belief Nets and Restricted...

In xrobin/DeepLearning: Deep Learning of neural networks

Description

Passes the data all the way through an unrolled DeepBeliefNet (in this case, it is identical to predict). For a RestrictedBolzmannMachine or a DeepBeliefNet that hasn't been unrolled, it will predict, and predict again through the reversed network. In the end, the reconstruction has the same dimension as the input.

Usage

reconstruct(object, newdata, ...)

## S3 method for class 'DeepBeliefNet'
reconstruct(object, newdata, drop = TRUE, ...)

## S3 method for class 'RestrictedBolzmannMachine'
reconstruct(object, newdata,
  drop = TRUE, ...)

Arguments

object	the RestrictedBolzmannMachine or DeepBeliefNet object
newdata	a data.frame or matrix providing the data. Must have the same columns than the input layer of the model.
...	ignored
drop	do not return additional dimensions

Value

the reconstructed data

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