Low-level R interface to train a LightGBM model. Unlike lightgbm, this function is focused on performance (e.g. speed, memory efficiency). It is also less likely to have breaking API changes in new releases than lightgbm.

  params = list(),
  nrounds = 100L,
  valids = list(),
  obj = NULL,
  eval = NULL,
  verbose = 1L,
  record = TRUE,
  eval_freq = 1L,
  init_model = NULL,
  colnames = NULL,
  categorical_feature = NULL,
  early_stopping_rounds = NULL,
  callbacks = list(),
  reset_data = FALSE,
  serializable = TRUE



a list of parameters. See the "Parameters" section of the documentation for a list of parameters and valid values.


a lgb.Dataset object, used for training. Some functions, such as lgb.cv, may allow you to pass other types of data like matrix and then separately supply label as a keyword argument.


number of training rounds


a list of lgb.Dataset objects, used for validation


objective function, can be character or custom objective function. Examples include regression, regression_l1, huber, binary, lambdarank, multiclass, multiclass


evaluation function(s). This can be a character vector, function, or list with a mixture of strings and functions.

  • a. character vector: If you provide a character vector to this argument, it should contain strings with valid evaluation metrics. See The "metric" section of the documentation for a list of valid metrics.

  • b. function: You can provide a custom evaluation function. This should accept the keyword arguments preds and dtrain and should return a named list with three elements:

    • name: A string with the name of the metric, used for printing and storing results.

    • value: A single number indicating the value of the metric for the given predictions and true values

    • higher_better: A boolean indicating whether higher values indicate a better fit. For example, this would be FALSE for metrics like MAE or RMSE.

  • c. list: If a list is given, it should only contain character vectors and functions. These should follow the requirements from the descriptions above.


verbosity for output, if <= 0 and valids has been provided, also will disable the printing of evaluation during training


Boolean, TRUE will record iteration message to booster$record_evals


evaluation output frequency, only effective when verbose > 0 and valids has been provided


path of model file or lgb.Booster object, will continue training from this model


feature names, if not null, will use this to overwrite the names in dataset


categorical features. This can either be a character vector of feature names or an integer vector with the indices of the features (e.g. c(1L, 10L) to say "the first and tenth columns").


int. Activates early stopping. When this parameter is non-null, training will stop if the evaluation of any metric on any validation set fails to improve for early_stopping_rounds consecutive boosting rounds. If training stops early, the returned model will have attribute best_iter set to the iteration number of the best iteration.


List of callback functions that are applied at each iteration.


Boolean, setting it to TRUE (not the default value) will transform the booster model into a predictor model which frees up memory and the original datasets


whether to make the resulting objects serializable through functions such as save or saveRDS (see section "Model serialization").


a trained booster model lgb.Booster.

Early Stopping

"early stopping" refers to stopping the training process if the model's performance on a given validation set does not improve for several consecutive iterations.

If multiple arguments are given to eval, their order will be preserved. If you enable early stopping by setting early_stopping_rounds in params, by default all metrics will be considered for early stopping.

If you want to only consider the first metric for early stopping, pass first_metric_only = TRUE in params. Note that if you also specify metric in params, that metric will be considered the "first" one. If you omit metric, a default metric will be used based on your choice for the parameter obj (keyword argument) or objective (passed into params).


# \donttest{
data(agaricus.train, package = "lightgbm")
train <- agaricus.train
dtrain <- lgb.Dataset(train$data, label = train$label)
data(agaricus.test, package = "lightgbm")
test <- agaricus.test
dtest <- lgb.Dataset.create.valid(dtrain, test$data, label = test$label)
params <- list(
  objective = "regression"
  , metric = "l2"
  , min_data = 1L
  , learning_rate = 1.0
  , num_threads = 2L
valids <- list(test = dtest)
model <- lgb.train(
  params = params
  , data = dtrain
  , nrounds = 5L
  , valids = valids
  , early_stopping_rounds = 3L
#> [LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.000930 seconds.
#> You can set `force_row_wise=true` to remove the overhead.
#> And if memory is not enough, you can set `force_col_wise=true`.
#> [LightGBM] [Info] Total Bins 232
#> [LightGBM] [Info] Number of data points in the train set: 6513, number of used features: 116
#> [LightGBM] [Info] Start training from score 0.482113
#> [LightGBM] [Warning] No further splits with positive gain, best gain: -inf
#> [1] "[1]:  test's l2:6.44165e-17"
#> [1] "Will train until there is no improvement in 3 rounds."
#> [LightGBM] [Warning] No further splits with positive gain, best gain: -inf
#> [1] "[2]:  test's l2:1.97215e-31"
#> [LightGBM] [Warning] No further splits with positive gain, best gain: -inf
#> [1] "[3]:  test's l2:0"
#> [LightGBM] [Warning] No further splits with positive gain, best gain: -inf
#> [LightGBM] [Warning] Stopped training because there are no more leaves that meet the split requirements
#> [1] "[4]:  test's l2:0"
#> [LightGBM] [Warning] No further splits with positive gain, best gain: -inf
#> [LightGBM] [Warning] Stopped training because there are no more leaves that meet the split requirements
#> [1] "[5]:  test's l2:0"
#> [1] "Did not meet early stopping, best iteration is: [3]:  test's l2:0"
# }