Philip Hyunsu Cho
Philip Hyunsu Cho
> there's a lot of packages that depend on xgboost in some way (currently 133 listed in CRAN alone), plus lots of scripts in the wild, and for a package...
@trivialfis We can host the R code in this repository, but we'd submit the code as `{xgboost2}` to CRAN. CRAN will still have the old XGBoost as `{xgboost}`.
I see. So we could define new functions suffixed with `Ex` or `ng` or something like that, if we want to keep using `{xgboost}` CRAN package. My personal preference is...
@trivialfis Also, we need to decide whether we want to try to submit the current XGBoost 2.0 to CRAN. My guess is that it would be a lot of work;...
> I'd say introducing a new interface in one go where everyone in this thread makes comments would make things easier, but it's more prone to bugs and hard to...
> I am wondering if it'd be possible to get helper features in the C interface from the core xgboost developers in order to improve the R interface. Yes, I...
According to #8352, `evals_result` was not part of the initial PySpark implementation. I am marking this as a feature request. @wbo4958 @WeichenXu123 How difficult do you think adding `evals_result` would...
I am the person who added the section `a-compatibility-note-for-saveRDS-save` to the manual, so let me answer the question regarding serialization. > How does the python scikit-learn interface deal with pickle...
The Cox objective might benefit from the same kind of regularization as the AFT objective, as described in Section 2.3 of https://arxiv.org/pdf/2006.04920.pdf It also might be reasonable to clip the...