mlr3torch
mlr3torch copied to clipboard
mlr-org
Input checking must be done before entering the private `.train`
mlr3torch would benefit from some additional input checks (e.g. whether the lazy tensor shape works with a specific learner), for that the mlr3::assert_task_learner() method would have to call the learner's input verification method.
The idea is, that such inconsistencies should not trigger a fallback learner, because it is a user error.
THis is a larger operation in mlr3, we can do it once it is solved there. closing for now