This PipeOp works for any regression learner.
The idea is to consider the whole task as a `Regression Task`

.
Hence, a numeric response vector is predicted for which crossvalidated rank thresholds are found using (nloptr)nloptr::nloptr to rank the predictions.
Returns a single `PredictionOrdinal`

.
As a default, optimizes `MeasureOrdinalCE`

.
Used for regression `Prediction`

s.

R6Class PipeOpOrdinalRegr

`R6::R6Class`

inheriting from `mlr3pipelines::PipeOpTaskPreproc`

.

Other PipeOps:
`mlr_pipeop_ordinalclassif`