Extreme Gradient Boosting Regression Learner

eXtreme Gradient Boosting regression. Calls xgboost::xgb.train() from package xgboost.

Custom mlr3 defaults

• nrounds:

  • Actual default: no default

  • Adjusted default: 1

  • Reason for change: Without a default construction of the learner would error. Just setting a nonsense default to workaround this. nrounds needs to be tuned by the user.

• verbose:

  • Actual default: 1

  • Adjusted default: 0

  • Reason for change: Reduce verbosity.

Dictionary

This Learner can be instantiated via the dictionary mlr_learners or with the associated sugar function lrn():

mlr_learners$get("regr.xgboost")
lrn("regr.xgboost")

References

Chen T, Guestrin C (2016). “Xgboost: A scalable tree boosting system.” In Proceedings of the 22nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 785--794. ACM. doi: 10.1145/2939672.2939785 .

See also

Dictionary of Learners: mlr3::mlr_learners

Super classes

mlr3::Learner -> mlr3::LearnerRegr -> LearnerRegrXgboost

Methods

Public methods

• LearnerRegrXgboost$new()

• LearnerRegrXgboost$importance()

• LearnerRegrXgboost$clone()

Inherited methods

• mlr3::Learner$format()

• mlr3::Learner$help()

• mlr3::Learner$predict()

• mlr3::Learner$predict_newdata()

• mlr3::Learner$print()

• mlr3::Learner$reset()

• mlr3::Learner$train()

---

Method new()

Creates a new instance of this R6 class.

Usage

LearnerRegrXgboost$new()

---

Method importance()

The importance scores are calculated with xgboost::xgb.importance().

Usage

LearnerRegrXgboost$importance()

Returns

Named numeric().

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerRegrXgboost$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

if (requireNamespace("xgboost")) {
  learner = mlr3::lrn("regr.xgboost")
  print(learner)

  # available parameters:
learner$param_set$ids()
}
#> <LearnerRegrXgboost:regr.xgboost>
#> * Model: -
#> * Parameters: nrounds=1, verbose=0
#> * Packages: xgboost
#> * Predict Type: response
#> * Feature types: logical, integer, numeric
#> * Properties: importance, missings, weights
#>  [1] "booster"                 "watchlist"              
#>  [3] "eta"                     "gamma"                  
#>  [5] "max_depth"               "min_child_weight"       
#>  [7] "subsample"               "colsample_bytree"       
#>  [9] "colsample_bylevel"       "colsample_bynode"       
#> [11] "num_parallel_tree"       "lambda"                 
#> [13] "lambda_bias"             "alpha"                  
#> [15] "objective"               "eval_metric"            
#> [17] "base_score"              "max_delta_step"         
#> [19] "missing"                 "monotone_constraints"   
#> [21] "tweedie_variance_power"  "nthread"                
#> [23] "nrounds"                 "feval"                  
#> [25] "verbose"                 "print_every_n"          
#> [27] "early_stopping_rounds"   "maximize"               
#> [29] "sample_type"             "normalize_type"         
#> [31] "rate_drop"               "skip_drop"              
#> [33] "one_drop"                "tree_method"            
#> [35] "grow_policy"             "max_leaves"             
#> [37] "max_bin"                 "callbacks"              
#> [39] "sketch_eps"              "scale_pos_weight"       
#> [41] "updater"                 "refresh_leaf"           
#> [43] "feature_selector"        "top_k"                  
#> [45] "predictor"               "save_period"            
#> [47] "save_name"               "xgb_model"              
#> [49] "interaction_constraints" "outputmargin"           
#> [51] "ntreelimit"              "predleaf"               
#> [53] "predcontrib"             "approxcontrib"          
#> [55] "predinteraction"         "reshape"                
#> [57] "training"