lightgbm: lgb.unloader – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

lgb.unloader

Remove lightgbm and its objects from an environment

Description

Attempts to unload LightGBM packages so you can remove objects cleanly without having to restart R. This is useful for instance if an object becomes stuck for no apparent reason and you do not want to restart R to fix the lost object.

Usage

lgb.unloader(restore = TRUE, wipe = FALSE, envir = .GlobalEnv)

Arguments

`restore`	Whether to reload `LightGBM` immediately after detaching from R. Defaults to `TRUE` which means automatically reload `LightGBM` once unloading is performed.
`wipe`	Whether to wipe all `lgb.Dataset` and `lgb.Booster` from the global environment. Defaults to `FALSE` which means to not remove them.
`envir`	The environment to perform wiping on if `wipe == TRUE`. Defaults to `.GlobalEnv` which is the global environment.

Value

NULL invisibly.

Examples

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")
valids <- list(test = dtest)
model <- lgb.train(
  params = params
  , data = dtrain
  , nrounds = 5L
  , valids = valids
  , min_data = 1L
  , learning_rate = 1.0
)

lgb.unloader(restore = FALSE, wipe = FALSE, envir = .GlobalEnv)
rm(model, dtrain, dtest) # Not needed if wipe = TRUE
gc() # Not needed if wipe = TRUE

library(lightgbm)
# Do whatever you want again with LightGBM without object clashing

lightgbm

Light Gradient Boosting Machine

v3.2.1

MIT + file LICENSE

Authors

Guolin Ke [aut, cre], Damien Soukhavong [aut], James Lamb [aut], Qi Meng [aut], Thomas Finley [aut], Taifeng Wang [aut], Wei Chen [aut], Weidong Ma [aut], Qiwei Ye [aut], Tie-Yan Liu [aut], Yachen Yan [ctb], Microsoft Corporation [cph], Dropbox, Inc. [cph], Jay Loden [cph], Dave Daeschler [cph], Giampaolo Rodola [cph], Alberto Ferreira [ctb], Daniel Lemire [ctb], Victor Zverovich [cph], IBM Corporation [ctb]

Initial release

2021-04-12