mb706
mb706
`learningrate.factor` should have a `$plus` and `$minus` component, `learningrate.limit` a `$min` and `$max` component.
Maybe when only one learner is used, preproc disregards permissible output types?
The error seems to be in buildWrapperSearchspace.R, where `dtWrap` contains `qNot(asQuoted(cbiparam))`, which evaluates to `TRUE` if there are no missings. Maybe the default value of `automlr.convert.before.impute` should be re-thought; otherwise,...
Trying to fix #2
(Sorry, this is a half-duplicate of #56, but I really think `coalesce` is bad and breaks code that uses it, even though it has the potential to be really useful.)...
```R > BBmisc::printHead(data.frame(as.list(1:10)), n = 5) X1L X2L X3L X4L X5L X6L X7L X8L X9L X10L 1 1 2 3 4 5 6 7 8 9 10 ... (#elements: 10)...
When `requirePackages` fails to load a package, it usually tells the user that the packages in question are not installed. This ignores the fact that packages may fail to load...
Are you planning to update the `kknn` version on CRAN, or is the GitHub version in some way the official release? Currently the invocation ```r replicate(100, kknn::kknn(speed ~ dist, cars[1:3,...
```r library("mlr3filters") flt("auc")$hash #> Error in mget(private$.extra_hash, envir = self) : invalid first argument ``` introduced by https://github.com/mlr-org/mlr3filters/pull/156 @mllg