bundle
bundle copied to clipboard
rstudio
Prepare objects for serialization with a consistent interface
I am not sure if this a known issue, as it doesn't appear in the [docs](https://rstudio.github.io/bundle/reference/bundle_h2o.html). It seems that except `predict`, other methods like `tidy` or `rank_results` fail using the...
https://github.com/rstudio/bundle/blob/29fd3ff1993a81f474d5a4990eac6d4fafe08068/R/bundle_xgboost.R#L42C22-L42C34 This argument was added in {xgboost} 1.6.0.1: https://github.com/cran/xgboost/commit/5cb303d85a65b55fe640fb581ca03a758c0fb662#diff-d0725fbcd63a81fecdbaaa1f592a6991840ee5bf9d13cb5b4747f1b6101e1ce6
I'm aware of the extended discussion in [#50](https://github.com/rstudio/bundle/issues/50) but I was still somewhat surprised that an xgboost model object after bundling & unbundling returns garbled variable names when trying to...
[tailor](https://tailor.tidymodels.org/) is a new (though not yet on CRAN) package from tidymodels that implements objects specifying postprocessing steps (analogous to recipes for preprocessing). Once that package is on CRAN, I'd...
We're seeing some unit test failures on Conda Forge in [a recent build for v0.1.2](https://dev.azure.com/conda-forge/feedstock-builds/_build/results?buildId=1078741&view=logs&j=7b6f2c87-f3a7-5133-8d84-7c03a75d9dfc&t=9eb77fd2-8ddd-5444-8fc0-71cb28dcb736&l=1206) when combined with `xgboost` v2.1.2. Root cause appears to be signature changes in `xgboost::xgb.DMatrix` and...
When using the unbundle function on a bundled XGBoost model, even after the 0.1.3 update, there is an error associated with the argument as_booster which has become deprecated. Below is...
Follow up on https://github.com/rstudio/vetiver-r/issues/300#issuecomment-3422969817 When the xgboost object `raw` vector is too long (>2^31), the unbundle function throws this error: ```R m 2**31 TRUE ``` In my case this happened...
I have performed a tuning process on a workflow set containing four workflows, which has taken 3-4 hours to complete. Because of the long processing time, I would like to...
