FTRL Proximal

This is an R package of the FTRL Proximal algorithm for online learning of elastic net logistic regression models.

For more info on the algorithm please see Ad Click Prediction: a View from the Trenches by McMahan et al. (2013).

Installation

Easiest way to install is from within R using the latest CRAN version:

install.packages("FTRLProximal")

If you want the latest build from git you can install it directly from github using devtools:

devtools::install_github("while/FTRLProximal")

Usage

Simplest use case is to use the model similar to normal glm with a model formula.

# Set up dataset
p <- mlbench::mlbench.2dnormals(100,2)
dat <- as.data.frame(p)

# Train model
mdl <- ftrlprox(classes ~ ., dat, lambda = 1e-2, alpha = 1, a = 0.3)

# Print resulting coeffs
print(mdl)

It is also possible to update the trained model object once it is trained.

# Set up first dataset
p <- mlbench.2dnormals(100,2)
dat <- as.data.frame(p)

# Convert data.frame to model.matrix
X <- model.matrix(classes ~ ., dat)

# Train on first dataset
mdl <- ftrlprox(X, dat$classes, lambda = 1e-2, alpha = 1, a = 0.3)

# Generate more of the same data after the first training session
p <- mlbench.2dnormals(100,2)
dat <- as.data.frame(p)

# Update model using the new data.
mdl <- update(mdl, X, dat$classes)

For more example please see the documentation.

Changelog

0.3

• Added prediction type "class".

0.2

• Changed from using explicit lambda1 and lambda2 parameters to using lambda and mixing parameter alpha.