Make caret results reproducible when a model trains in parallel on Windows or Linux, by setting the random seeds caret uses.

Turn an analysis into an R package, with the treebgg bagging package as the worked example, covering roxygen2 documents, testthat tests and vignettes.

Compare the snow and foreach approaches to parallel processing in R on three practical examples, starting with k-means clustering on the Boston data.

Loop in parallel on one machine in R with the foreach and doParallel packages, with the iterators package covered for writing the loop itself.

Run parallel tasks on one machine in R with the snow and parallel packages, comparing clusterApply, clusterApplyLB, parLapply and parLapplyLB.

Compare a single classification tree in R with 500 bagged trees on out-of-bag and test errors, cumulative errors and variable importance measures.

Evaluate a single regression tree in R against 2000 bagged trees, comparing out-of-bag and test errors, cumulative errors and variable importance.

Compare three R packages for classification trees, rpart, caret and mlr, on the Carseats data used in the earlier parts.

Fit a regression tree on the Carseats data in R, comparing the pruning parameter caret selects against the 1-SE rule that the rpart package recommends.

Cost-sensitive classification with rpart and caret in R, treating a missed High class as twice as expensive by altering the priors and the loss matrix.