Subclass for random search tuning.
The random points are sampled by paradox::generate_design_random().
Source
Bergstra J, Bengio Y (2012). “Random Search for Hyper-Parameter Optimization.” Journal of Machine Learning Research, 13(10), 281--305. https://jmlr.csail.mit.edu/papers/v13/bergstra12a.html.
Dictionary
This Tuner can be instantiated via the dictionary
mlr_tuners or with the associated sugar function tnr():
TunerRandomSearch$new()
mlr_tuners$get("random_search")
tnr("random_search")Parallelization
In order to support general termination criteria and parallelization, we
evaluate points in a batch-fashion of size batch_size. Larger batches mean
we can parallelize more, smaller batches imply a more fine-grained checking
of termination criteria. A batch contains of batch_size times resampling$iters jobs.
E.g., if you set a batch size of 10 points and do a 5-fold cross validation, you can
utilize up to 50 cores.
Parallelization is supported via package future (see mlr3::benchmark()'s
section on parallelization for more details).
Logging
All Tuners use a logger (as implemented in lgr) from package
bbotk.
Use lgr::get_logger("bbotk") to access and control the logger.
Progress Bars
$optimize() supports progress bars via the package progressr
combined with a Terminator. Simply wrap the function in
progressr::with_progress() to enable them. We recommend to use package
progress as backend; enable with progressr::handlers("progress").
See also
Package mlr3hyperband for hyperband tuning.
Other Tuner:
mlr_tuners_cmaes,
mlr_tuners_design_points,
mlr_tuners_gensa,
mlr_tuners_grid_search,
mlr_tuners_irace,
mlr_tuners_nloptr
Super classes
mlr3tuning::Tuner -> mlr3tuning::TunerFromOptimizer -> TunerRandomSearch
Examples
# retrieve task
task = tsk("pima")
# load learner and set search space
learner = lrn("classif.rpart", cp = to_tune(1e-04, 1e-1, logscale = TRUE))
# hyperparameter tuning on the pima indians diabetes data set
instance = tune(
method = "random_search",
task = task,
learner = learner,
resampling = rsmp("holdout"),
measure = msr("classif.ce"),
term_evals = 10
)
# best performing hyperparameter configuration
instance$result
#> cp learner_param_vals x_domain classif.ce
#> 1: -7.133431 <list[2]> <list[1]> 0.234375
# all evaluated hyperparameter configuration
as.data.table(instance$archive)
#> cp classif.ce x_domain_cp runtime_learners timestamp
#> 1: -7.133431 0.234375 0.0007979770 0.017 2022-01-23 04:26:46
#> 2: -8.600755 0.234375 0.0001839669 0.015 2022-01-23 04:26:46
#> 3: -8.106442 0.234375 0.0003015899 0.015 2022-01-23 04:26:46
#> 4: -6.541400 0.234375 0.0014424673 0.018 2022-01-23 04:26:46
#> 5: -8.363449 0.234375 0.0002332386 0.015 2022-01-23 04:26:46
#> 6: -7.966170 0.234375 0.0003470054 0.018 2022-01-23 04:26:47
#> 7: -6.133246 0.234375 0.0021695275 0.016 2022-01-23 04:26:47
#> 8: -7.932124 0.234375 0.0003590231 0.015 2022-01-23 04:26:47
#> 9: -5.003379 0.234375 0.0067152196 0.017 2022-01-23 04:26:47
#> 10: -9.111880 0.234375 0.0001103471 0.016 2022-01-23 04:26:47
#> batch_nr resample_result
#> 1: 1 <ResampleResult[22]>
#> 2: 2 <ResampleResult[22]>
#> 3: 3 <ResampleResult[22]>
#> 4: 4 <ResampleResult[22]>
#> 5: 5 <ResampleResult[22]>
#> 6: 6 <ResampleResult[22]>
#> 7: 7 <ResampleResult[22]>
#> 8: 8 <ResampleResult[22]>
#> 9: 9 <ResampleResult[22]>
#> 10: 10 <ResampleResult[22]>
# fit final model on complete data set
learner$param_set$values = instance$result_learner_param_vals
learner$train(task)