[R-package] remove support for '...' in lgb.cv()

R-package/R/lgb.cv.R

@@ -43,18 +43,6 @@ CVBooster <- R6::R6Class(
 #' @param callbacks List of callback functions that are applied at each iteration.
 #' @param reset_data 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
-#' @param ... other parameters, see Parameters.rst for more information. A few key parameters:
-#'            \itemize{
-#'                \item{\code{boosting}: Boosting type. \code{"gbdt"}, \code{"rf"}, \code{"dart"} or \code{"goss"}.}
-#'                \item{\code{num_leaves}: Maximum number of leaves in one tree.}
-#'                \item{\code{max_depth}: Limit the max depth for tree model. This is used to deal with
-#'                                 overfit when #data is small. Tree still grow by leaf-wise.}
-#'                \item{\code{num_threads}: Number of threads for LightGBM. For the best speed, set this to
-#'                             the number of real CPU cores(\code{parallel::detectCores(logical = FALSE)}),
-#'                             not the number of threads (most CPU using hyper-threading to generate 2 threads
-#'                             per CPU core).}
-#'            }
-#'            NOTE: As of v3.3.0, use of \code{...} is deprecated. Add parameters to \code{params} directly.
 #' @inheritSection lgb_shared_params Early Stopping
 #' @return a trained model \code{lgb.CVBooster}.
 #'
@@ -99,7 +87,6 @@ lgb.cv <- function(params = list()
                    , callbacks = list()
                    , reset_data = FALSE
                    , serializable = TRUE
-                   , ...
                    ) {
 
   if (nrounds <= 0L) {
@@ -115,23 +102,12 @@ lgb.cv <- function(params = list()
   }
 
   # Setup temporary variables
-  additional_params <- list(...)
-  params <- append(params, additional_params)
   params$verbose <- verbose
   params <- lgb.check.obj(params = params, obj = obj)
   params <- lgb.check.eval(params = params, eval = eval)
   fobj <- NULL
   eval_functions <- list(NULL)
 
-  if (length(additional_params) > 0L) {
-    warning(paste0(
-      "lgb.cv: Found the following passed through '...': "
-      , paste(names(additional_params), collapse = ", ")
-      , ". These will be used, but in future releases of lightgbm, this warning will become an error. "
-      , "Add these to 'params' instead. See ?lgb.cv for documentation on how to call this function."
-    ))
-  }
-
   # set some parameters, resolving the way they were passed in with other parameters
   # in `params`.
   # this ensures that the model stored with Booster$save() correctly represents

R-package/tests/testthat/test_basic.R

@@ -347,30 +347,36 @@ context("lgb.cv()")
 
 test_that("cv works", {
   dtrain <- lgb.Dataset(train$data, label = train$label)
-  params <- list(objective = "regression", metric = "l2,l1")
+  params <- list(
+    objective = "regression"
+    , metric = "l2,l1"
+    , min_data = 1L
+    , learning_rate = 1.0
+  )
   bst <- lgb.cv(
     params
     , dtrain
     , 10L
     , nfold = 5L
-    , min_data = 1L
-    , learning_rate = 1.0
     , early_stopping_rounds = 10L
   )
   expect_false(is.null(bst$record_evals))
 })
 
 test_that("lgb.cv() rejects negative or 0 value passed to nrounds", {
   dtrain <- lgb.Dataset(train$data, label = train$label)
-  params <- list(objective = "regression", metric = "l2,l1")
+  params <- list(
+    objective = "regression"
+    , metric = "l2,l1"
+    , min_data = 1L
+  )
   for (nround_value in c(-10L, 0L)) {
     expect_error({
       bst <- lgb.cv(
         params
         , dtrain
         , nround_value
         , nfold = 5L
-        , min_data = 1L
       )
     }, "nrounds should be greater than zero")
   }
@@ -388,11 +394,14 @@ test_that("lgb.cv() throws an informative error is 'data' is not an lgb.Dataset
   for (val in bad_values) {
     expect_error({
       bst <- lgb.cv(
-        params = list(objective = "regression", metric = "l2,l1")
+        params = list(
+            objective = "regression"
+            , metric = "l2,l1"
+            , min_data = 1L
+        )
         , data = val
         , 10L
         , nfold = 5L
-        , min_data = 1L
       )
     }, regexp = "'label' must be provided for lgb.cv if 'data' is not an 'lgb.Dataset'", fixed = TRUE)
   }
@@ -409,11 +418,11 @@ test_that("lightgbm.cv() gives the correct best_score and best_iter for a metric
     data = dtrain
     , nfold = 5L
     , nrounds = nrounds
-    , num_leaves = 5L
     , params = list(
       objective = "binary"
       , metric = "auc,binary_error"
       , learning_rate = 1.5
+      , num_leaves = 5L
     )
   )
   expect_is(cv_bst, "lgb.CVBooster")
@@ -470,15 +479,18 @@ test_that("lgb.cv() fit on linearly-relatead data improves when using linear lea
 
 test_that("lgb.cv() respects showsd argument", {
   dtrain <- lgb.Dataset(train$data, label = train$label)
-  params <- list(objective = "regression", metric = "l2")
+  params <- list(
+    objective = "regression"
+    , metric = "l2"
+    , min_data = 1L
+  )
   nrounds <- 5L
   set.seed(708L)
   bst_showsd <- lgb.cv(
     params = params
     , data = dtrain
     , nrounds = nrounds
     , nfold = 3L
-    , min_data = 1L
     , showsd = TRUE
   )
   evals_showsd <- bst_showsd$record_evals[["valid"]][["l2"]]
@@ -488,7 +500,6 @@ test_that("lgb.cv() respects showsd argument", {
     , data = dtrain
     , nrounds = nrounds
     , nfold = 3L
-    , min_data = 1L
     , showsd = FALSE
   )
   evals_no_showsd <- bst_no_showsd$record_evals[["valid"]][["l2"]]

R-package/tests/testthat/test_learning_to_rank.R

@@ -78,6 +78,8 @@ test_that("learning-to-rank with lgb.cv() works as expected", {
         , ndcg_at = ndcg_at
         , lambdarank_truncation_level = 3L
         , label_gain = "0,1,3"
+        , min_data = 1L
+        , learning_rate = 0.01
     )
     nfold <- 4L
     nrounds <- 10L
@@ -86,8 +88,6 @@ test_that("learning-to-rank with lgb.cv() works as expected", {
         , data = dtrain
         , nrounds = nrounds
         , nfold = nfold
-        , min_data = 1L
-        , learning_rate = 0.01
     )
     expect_is(cv_bst, "lgb.CVBooster")
     expect_equal(length(cv_bst$boosters), nfold)