Some libraries support multiple outputs from one tree / forest e.g. https://xgboost.readthedocs.io/en/stable/tutorials/multioutput.html
This would be good to support in conifer, especially as the conventional approach to multiple output targets (e.g. multi-class classification) involves adding more trees (n_learners * n_classes). Multiple output models ought to be more resource efficient.
Some libraries support multiple outputs from one tree / forest e.g. https://xgboost.readthedocs.io/en/stable/tutorials/multioutput.html
This would be good to support in
conifer, especially as the conventional approach to multiple output targets (e.g. multi-class classification) involves adding more trees (n_learners * n_classes). Multiple output models ought to be more resource efficient.