ensemble: Ensemble Learning to extend single model
In SkadiEye/ITRlearn: Individualized Treatment Rule Learning

Description Usage Arguments Value See Also

Ensemble Learning to extend a reg, wcls or a cvGrid method. Multiple models will be fitted using randomly generated bootstrap samples. The out-of-bag (OOB) samples will be used as the validation set if the corresponding method could use the validation to fit the model. The predictions from all models will be aggregated when predicting for new data. Each model will be evaluated and assigned a performance score.

1
2
3

ensemble(object, n.ensemble = 100, cutoff = 0.5, esCtrl,
  unbalance.trt = c("None", "Over", "Under", "Balance")[1], prop.keep = 1,
  best.opti = TRUE, min.keep = 10, parallel = FALSE, n.nodes = 4)

`object`	A `RegObj` or `wClsObj` containing the training set.
`n.ensemble`	The number of models in the ensemble.
`cutoff`	Cutoff used when predicting OOB samples, passed to `cutoff.calc` function.
`esCtrl`	A `list` of arguments that specifies a regression or classification method, handled by `esCtrlPanel`.
`unbalance.trt`	A character string that specifies if sampleing techniques should be applied to data with unbalanced numbers for each class.
`prop.keep`	A `numeric` value that specifies the percentage of models to keep with top performance scores.
`best.opti`	A `logical` value, whether the number of model to keep is determined automatically.
`min.keep`	A `numeric` value that specifies the minimum models to keep.