train.xgboost: train.xgboost

In traineR: Predictive (Classification and Regression) Models Homologator

train.xgboost

Description

This function wraps xgb.train to standardize model training within the traineR framework. It automatically handles preprocessing, parameter configuration, multiclass settings, and metadata generation for predictions.

Usage

train.xgboost(
  formula,
  data,
  nrounds,
  evals = list(),
  custom_metric = NULL,
  verbose = 1,
  print_every_n = 1L,
  early_stopping_rounds = NULL,
  maximize = NULL,
  save_period = NULL,
  save_name = "xgboost.model",
  xgb_model = NULL,
  callbacks = list(),
  eval_metric = NULL,
  extra_params = NULL,
  booster = "gbtree",
  objective = NULL,
  eta = 0.3,
  gamma = 0,
  max_depth = 6,
  min_child_weight = 1,
  subsample = 1,
  colsample_bytree = 1,
  ...
)

Arguments

formula	A model formula describing the response and predictors.
data	A data frame containing the training data. Internally, it is converted to an xgb.DMatrix.
nrounds	Maximum number of boosting iterations.
evals	A named list of xgb.DMatrix objects for evaluation during training. Defaults to training data if empty.
custom_metric	A custom evaluation function for xgboost.
verbose	Controls verbosity: 0 = silent, 1 = progress printed.
print_every_n	Print evaluation results every print_every_n iterations.
early_stopping_rounds	Number of rounds with no improvement before stopping.
maximize	Logical indicating if the evaluation metric should be maximized.
save_period	Save the model every save_period rounds. Defaults to saving at the end.
save_name	File name for saving the model.
xgb_model	A previously trained xgboost model for continuation.
callbacks	A list of callback functions for xgboost during training.
eval_metric	Evaluation metric for xgboost (e.g., "mlogloss", "rmse").
extra_params	Optional list of additional xgboost parameters.
booster	Booster type: "gbtree" or "gblinear". Default is "gbtree".
objective	Objective function for xgboost. If NULL, it's chosen automatically: Regression → "reg:squarederror" Binary classification → "binary:logistic" Multiclass → "multi:softprob"
eta	Learning rate. Default is 0.3.
gamma	Minimum loss reduction for a split. Default is 0.
max_depth	Maximum depth of trees. Default is 6.
min_child_weight	Minimum sum of instance weight in a child.
subsample	Subsample ratio for training instances. Default is 1.
colsample_bytree	Subsample ratio of columns per tree. Default is 1.
...	Additional arguments for xgb.train.

Value

An object of class xgb.Booster.prmdt containing:

• The trained xgboost model.

• Metadata used by traineR for prediction output.

See Also

xgb.train, xgb.DMatrix

traineR documentation built on Jan. 17, 2026, 9:07 a.m.