Rnn (software)

rnn is an open-source machine learning framework that implements recurrent neural network architectures, such as LSTM and GRU, natively in the R programming language, that has been downloaded over 100,000 times (from the RStudio servers alone).^[1]

rnn

Original author(s)	Bastiaan Quast
Initial release	30 November 2015 (2015-11-30)
Stable release	1.9.0 / 22 April 2023; 23 months ago (2023-04-22)
Preview release	1.9.0.9000 / 22 April 2023; 23 months ago (2023-04-22)
Repository	github.com/bquast/rnn
Written in	R
Operating system	macOS, Linux, Windows
Size	564.2 kB (v. 1.9.0)
License	GPL v3
Website	cran.r-project.org/web/packages/rnn/

The rnn package is distributed through the Comprehensive R Archive Network^[2] under the open-source GPL v3 license.

Workflow

Demonstration of RNN package

The below example from the rnn documentation show how to train a recurrent neural network to solve the problem of bit-by-bit binary addition.

> # install the rnn package, including the dependency sigmoid
> install.packages('rnn')

> # load the rnn package
> library(rnn)

> # create input data 
> X1 = sample(0:127, 10000, replace=TRUE)
> X2 = sample(0:127, 10000, replace=TRUE)

> # create output data
> Y <- X1 + X2

> # convert from decimal to binary notation 
> X1 <- int2bin(X1, length=8)
> X2 <- int2bin(X2, length=8)
> Y  <- int2bin(Y,  length=8)

> # move input data into single tensor
> X <- array( c(X1,X2), dim=c(dim(X1),2) )

> # train the model
> model <- trainr(Y=Y,
+                 X=X,
+                 learningrate   =  1,
+                 hidden_dim     = 16  )
Trained epoch: 1 - Learning rate: 1
Epoch error: 0.839787019539748

sigmoid

The sigmoid functions and derivatives used in the package were originally included in the package, from version 0.8.0 onwards, these were released in a separate R package sigmoid, with the intention to enable more general use. The sigmoid package is a dependency of the rnn package and therefore automatically installed with it.^[3]

Reception

With the release of version 0.3.0 in April 2016^[4] the use in production and research environments became more widespread. The package was reviewed several months later on the R blog The Beginner Programmer as "R provides a simple and very user friendly package named rnn for working with recurrent neural networks.",^[5] which further increased usage.^[6]

The book Neural Networks in R by Balaji Venkateswaran and Giuseppe Ciaburro uses rnn to demonstrate recurrent neural networks to R users.^[7][8] It is also used in the r-exercises.com course "Neural network exercises".^[9][10]

The RStudio CRAN mirror download logs ^[11] show that the package is downloaded on average about 2,000 per month from those servers ,^[12] with a total of over 100,000 downloads since the first release,^[13] according to RDocumentation.org, this puts the package in the 15th percentile of most popular R packages .^[14]