The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

Parallelize 'DiceKriging' functions

The 'futurize' hexlogo = The 'future' logo

The futurize package allows you to easily turn sequential code into parallel code by piping the sequential code to the futurize() function. Easy!

TL;DR

library(futurize)
plan(multisession)
library(DiceKriging)

design <- expand.grid(x1 = seq(0, 1, length = 15), x2 = seq(0, 1, length = 15))
y <- apply(design, 1, function(x) x[1]^2 + x[2]^2)
m <- km(~., design = design, response = data.frame(y = y),
        multistart = 20) |> futurize()

Introduction

This vignette demonstrates how to use futurize to parallelize DiceKriging functions, specifically km(). When fitting a kriging model via km(), the parameters of the covariance function are estimated by maximum likelihood or cross-validation. The optimization can be started from multiple points (to avoid local optima), which can be done in parallel.

Example: kriging model with multi-start optimization

Fitting a kriging model with a single starting point:

library(DiceKriging)

design <- expand.grid(x1 = seq(0, 1, length = 15), x2 = seq(0, 1, length = 15))
y <- apply(design, MARGIN = 1, FUN = function(x) x[1]^2 + x[2]^2)
m <- km(~., design = design, response = data.frame(y = y))

To run multiple optimizer starts in parallel, set multistart > 1 and pipe to futurize():

library(futurize)
library(DiceKriging)

design <- expand.grid(x1 = seq(0, 1, length = 15), x2 = seq(0, 1, length = 15))
y <- apply(design, MARGIN = 1, FUN = function(x) x[1]^2 + x[2]^2)
m <- km(~., design = design, response = data.frame(y = y),
        multistart = 20) |> futurize()

This distributes the multi-start runs across the available parallel workers, given that we have set up a parallel plan, e.g.

plan(multisession)

The built-in multisession backend parallelizes on your local computer and works on all operating systems. There are other parallel backends to choose from, including alternatives to parallelize locally as well as distributed across remote machines, e.g.

plan(future.mirai::mirai_multisession)

and

plan(future.batchtools::batchtools_slurm)

Supported Functions

The following DiceKriging function is supported by futurize():

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.