Tidymodels & XGBoost & evaluation.log

[John_Rambo]

With:

myControl<-control_grid(
                save_workflow  = TRUE
                ,save_pred      = TRUE
                 ,extract = function(x) {
                     model <- extract_fit_engine(x)
                     model$evaluation_log
                
                 }
            )

... we can get via:

                pred_errors<-xgb_tuning_results %>%
                    collect_extracts() %>%
                    tidyr::unnest(.extracts)

validation errors OR training errors from xgboost.

I wondering if it is possible to get both at the same time- like in native xgboost.
Any ideas here? Is there a hack via environment?