Mastering Machine Learning with R - Second Edition

Random forest

With our training data task created, you have a number of functions to explore it. Here is the abbreviated output that looks at its structure:

    > str(getTaskData(wine.task))    'data.frame': 438 obs. of 14 variables:     $ class: Factor w/ 3 levels "1","2","3": 1 2 1 2 2 1 2 1 1 2 ...     $ V1 : num 13.6 11.8 14.4 11.8 13.1 ...

There are many ways to use mlr in your analysis, but I recommend creating your resample object. Here we create a resampling object to help us in tuning the number of trees for our random forest, consisting of three subsamples:

    > rdesc <- makeResampleDesc("Subsample", iters = 3)

The next object establishes the grid of trees for tuning with the minimum number of trees, 750, and the maximum of 2000. You can also ...