mlr3tuning

Package website: release | dev

r-cmd-check CRAN Status StackOverflow Mattermost

mlr3tuning is the hyperparameter optimization package of the mlr3 ecosystem. It features highly configurable search spaces via the paradox package and finds optimal hyperparameter configurations for any mlr3 learner. mlr3tuning works with several optimization algorithms e.g. Random Search, Iterated Racing, Bayesian Optimization (in mlr3mbo) and Hyperband (in mlr3hyperband). Moreover, it can automatically optimize learners and estimate the performance of optimized models with nested resampling. The package is built on the optimization framework bbotk.

Extension packages

mlr3tuning is extended by the following packages.

Resources

There are several sections about hyperparameter optimization in the mlr3book.

The gallery features a collection of case studies and demos about optimization.

The cheatsheet summarizes the most important functions of mlr3tuning.

Installation

Install the last release from CRAN:

install.packages("mlr3tuning")

Install the development version from GitHub:

remotes::install_github("mlr-org/mlr3tuning")

Examples

We optimize the cost and gamma hyperparameters of a support vector machine on the Sonar data set.

library("mlr3learners")
library("mlr3tuning")

learner = lrn("classif.svm",
  cost  = to_tune(1e-5, 1e5, logscale = TRUE),
  gamma = to_tune(1e-5, 1e5, logscale = TRUE),
  kernel = "radial",
  type = "C-classification"
)

We construct a tuning instance with the ti() function. The tuning instance describes the tuning problem.

instance = ti(
  task = tsk("sonar"),
  learner = learner,
  resampling = rsmp("cv", folds = 3),
  measures = msr("classif.ce"),
  terminator = trm("none")
)
instance
## <TuningInstanceBatchSingleCrit>
## * State:  Not optimized
## * Objective: <ObjectiveTuningBatch:classif.svm_on_sonar>
## * Search Space:
##       id    class     lower    upper nlevels
## 1:  cost ParamDbl -11.51293 11.51293     Inf
## 2: gamma ParamDbl -11.51293 11.51293     Inf
## * Terminator: <TerminatorNone>

We select a simple grid search as the optimization algorithm.

tuner = tnr("grid_search", resolution = 5)
tuner
## <TunerBatchGridSearch>: Grid Search
## * Parameters: batch_size=1, resolution=5
## * Parameter classes: ParamLgl, ParamInt, ParamDbl, ParamFct
## * Properties: dependencies, single-crit, multi-crit
## * Packages: mlr3tuning, bbotk

To start the tuning, we simply pass the tuning instance to the tuner.

tuner$optimize(instance)
##        cost     gamma learner_param_vals  x_domain classif.ce
## 1: 5.756463 -5.756463          <list[4]> <list[2]>  0.2063492

The tuner returns the best hyperparameter configuration and the corresponding measured performance.

The archive contains all evaluated hyperparameter configurations.

as.data.table(instance$archive)[, .(cost, gamma, classif.ce, batch_nr, resample_result)]
##           cost     gamma classif.ce batch_nr  resample_result
##  1:   5.756463  0.000000  0.4661146        1 <ResampleResult>
##  2:  11.512925  0.000000  0.4661146        2 <ResampleResult>
##  3: -11.512925  5.756463  0.4661146        3 <ResampleResult>
##  4: -11.512925  0.000000  0.4661146        4 <ResampleResult>
##  5:  -5.756463 11.512925  0.4661146        5 <ResampleResult>
## ---                                                          
## 21:  11.512925  5.756463  0.4661146       21 <ResampleResult>
## 22: -11.512925 11.512925  0.4661146       22 <ResampleResult>
## 23:   0.000000  5.756463  0.4661146       23 <ResampleResult>
## 24:   5.756463 11.512925  0.4661146       24 <ResampleResult>
## 25: -11.512925 -5.756463  0.4661146       25 <ResampleResult>

The mlr3viz package visualizes tuning results.

library(mlr3viz)

autoplot(instance, type = "surface")

We fit a final model with optimized hyperparameters to make predictions on new data.

learner$param_set$values = instance$result_learner_param_vals
learner$train(tsk("sonar"))