Hannah Frick
Hannah Frick
We currently don't do any "home-grown" type checking in censored. But maybe we should add type-checking.
I noticed that flexsurv's predict method does not nest the survival probabilities if there's only one time point to calculate the probability at. While nesting isn't neccessary in that case...
I'm seeing a lot of warnings in the tests of the type below, would be nice to clean up 🧹 ``` Warning (test-mars.R:72:3): check model reduction non-uniform 'Rounding' sampler used...
It should also be listed in the `@description` https://github.com/tidymodels/hardhat/blob/ae8fba73a703a5d5123e67f3fcbeb4ebbe6a8825/R/extract.R#L3-L17
The steps `step_mutate_at()` and `step_rename_at()` are based on superseded dplyr functions `mutate_at()` and `rename_at()`.