Takes a lists of Task, a list of Learner and a list of Resampling to
generate a design in an expand.grid() fashion (a.k.a. cross join or Cartesian product).
Resampling strategies are not allowed to be instantiated when passing the argument, and instead will be instantiated per task internally.
benchmark_grid(tasks, learners, resamplings)
| tasks | :: list of Task. |
|---|---|
| learners | :: list of Learner. |
| resamplings | :: list of Resampling. |
(data.table::data.table()) with the cross product of the input vectors.
tasks = list(tsk("iris"), tsk("sonar")) learners = list(lrn("classif.featureless"), lrn("classif.rpart")) resamplings = list(rsmp("cv"), rsmp("subsampling")) benchmark_grid(tasks, learners, resamplings)#> task learner resampling #> 1: <TaskClassif> <LearnerClassifFeatureless> <ResamplingCV> #> 2: <TaskClassif> <LearnerClassifFeatureless> <ResamplingSubsampling> #> 3: <TaskClassif> <LearnerClassifRpart> <ResamplingCV> #> 4: <TaskClassif> <LearnerClassifRpart> <ResamplingSubsampling> #> 5: <TaskClassif> <LearnerClassifFeatureless> <ResamplingCV> #> 6: <TaskClassif> <LearnerClassifFeatureless> <ResamplingSubsampling> #> 7: <TaskClassif> <LearnerClassifRpart> <ResamplingCV> #> 8: <TaskClassif> <LearnerClassifRpart> <ResamplingSubsampling>