makeMobafeasObjective: Create mlrMBO Objective Function
In compstat-lmu/mobafeas: Model Based Feature Selection

Description Usage Arguments Value See Also

Create an objective function that resamples learner on task with resampling and measures measure (optional), together with the number of features selected.

The ParamSet given should only apply to the learner (and the cpo if given); a parameter selector.selection is added to it. The complete parameter set of the function can be obtained using getParamSet().

learner must not include a cpoSelector() applied to it, this happens automatically within makeObjective.

1
2
3

makeMobafeasObjective(learner, task, ps = pSS(), resampling,
  measure = NULL, holdout.data = NULL, worst.measure = NULL,
  cpo = NULLCPO, multi.objective = TRUE)

`learner`	`[Learner]` A `Learner` object to optimize.
`task`	`[Task]` The `mlr::Task` object to optimize on.
`ps`	`[ParamSet]` The `ParamSet` to optimize over.
`resampling`	`[ResampleDesc \| ResampleInst]` The `ResampleDesc` or `ResampleInst` object to use.
`measure`	`[Measure \| NULL]` The `Measure` to optimize for. The default is `NULL`, which uses the `task`'s default `Measure`.
`holdout.data`	`[Task]` Additional data on which to predict each configuration after training on `task`.
`worst.measure`	`[numeric(1)]` worst value for measure to consider, for dominated hypervolume calculation. Will be extracted from the given measure if not given, but will raise an error if the extracted (or given) value is infinite.
`cpo`	`[CPO]` CPO pipeline to apply before feature selection. (A CPO that should be applied after feature selection should already be part of `learner` when given). Care should be taken that the `selector.selection` parameter in `ps` has the appropriate length of the data that `cpo` emits.
`multi.objective`	`[logical(1) \| function]` Whether to create a multi-objective function. Default `TRUE`. `FALSE`: simulate single-objective optimization by setting the second-objective value to 0 by default. This may also be a function with two inputs `performance`, `featfrac` that computes a scalarized objective. The scalarized objective must always be minimized.