# nolint start
library(mlexperiments)
library(mllrnrs)
# nolint start
library(mlexperiments)
library(mllrnrs)
See https://github.com/kapsner/mllrnrs/blob/main/R/learner_lightgbm.R for implementation details.
library(mlbench)
data("DNA")
<- DNA |>
dataset ::as.data.table() |>
data.tablena.omit()
<- colnames(dataset)[1:180]
feature_cols <- "Class" target_col
<- 123
seed if (isTRUE(as.logical(Sys.getenv("_R_CHECK_LIMIT_CORES_")))) {
# on cran
<- 2L
ncores else {
} <- ifelse(
ncores test = parallel::detectCores() > 4,
yes = 4L,
no = ifelse(
test = parallel::detectCores() < 2L,
yes = 1L,
no = parallel::detectCores()
)
)
}options("mlexperiments.bayesian.max_init" = 10L)
options("mlexperiments.optim.lgb.nrounds" = 100L)
options("mlexperiments.optim.lgb.early_stopping_rounds" = 10L)
<- splitTools::partition(
data_split y = dataset[, get(target_col)],
p = c(train = 0.7, test = 0.3),
type = "stratified",
seed = seed
)
<- model.matrix(
train_x ~ -1 + .,
$train, .SD, .SDcols = feature_cols]
dataset[data_split
)<- as.integer(dataset[data_split$train, get(target_col)]) - 1L
train_y
<- model.matrix(
test_x ~ -1 + .,
$test, .SD, .SDcols = feature_cols]
dataset[data_split
)<- as.integer(dataset[data_split$test, get(target_col)]) - 1L test_y
<- splitTools::create_folds(
fold_list y = train_y,
k = 3,
type = "stratified",
seed = seed
)
# required learner arguments, not optimized
<- list(
learner_args max_depth = -1L,
verbose = -1L,
objective = "multiclass",
metric = "multi_logloss",
num_class = "3"
)
# set arguments for predict function and performance metric,
# required for mlexperiments::MLCrossValidation and
# mlexperiments::MLNestedCV
<- list(reshape = TRUE)
predict_args <- metric("bacc")
performance_metric <- NULL
performance_metric_args <- FALSE
return_models
# required for grid search and initialization of bayesian optimization
<- expand.grid(
parameter_grid bagging_fraction = seq(0.6, 1, .2),
feature_fraction = seq(0.6, 1, .2),
min_data_in_leaf = seq(2, 10, 2),
learning_rate = seq(0.1, 0.2, 0.1),
num_leaves = seq(2, 20, 4)
)# reduce to a maximum of 10 rows
if (nrow(parameter_grid) > 10) {
set.seed(123)
<- sample(seq_len(nrow(parameter_grid)), 10, FALSE)
sample_rows <- kdry::mlh_subset(parameter_grid, sample_rows)
parameter_grid
}
# required for bayesian optimization
<- list(
parameter_bounds bagging_fraction = c(0.2, 1),
feature_fraction = c(0.2, 1),
min_data_in_leaf = c(2L, 12L),
learning_rate = c(0.1, 0.2),
num_leaves = c(2L, 20L)
)<- list(
optim_args iters.n = ncores,
kappa = 3.5,
acq = "ucb"
)
<- mlexperiments::MLTuneParameters$new(
tuner learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),strategy = "grid",
ncores = ncores,
seed = seed
)
$parameter_grid <- parameter_grid
tuner$learner_args <- learner_args
tuner$split_type <- "stratified"
tuner
$set_data(
tunerx = train_x,
y = train_y
)
<- tuner$execute(k = 3)
tuner_results_grid #> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [==================>-----------------------------------------------------------------------------] 2/10 ( 20%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [============================>-------------------------------------------------------------------] 3/10 ( 30%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [=====================================>----------------------------------------------------------] 4/10 ( 40%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [===============================================>------------------------------------------------] 5/10 ( 50%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [=========================================================>--------------------------------------] 6/10 ( 60%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [==================================================================>-----------------------------] 7/10 ( 70%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [============================================================================>-------------------] 8/10 ( 80%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [=====================================================================================>----------] 9/10 ( 90%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#>
#> Parameter settings [===============================================================================================] 10/10 (100%)
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -1.424655
#> [LightGBM] [Info] Start training from score -0.656204
#> [LightGBM] [Info] Start training from score -1.422637
#> [LightGBM] [Info] Start training from score -1.428239
#> [LightGBM] [Info] Start training from score -0.655482
head(tuner_results_grid)
#> setting_id metric_optim_mean nrounds bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves max_depth
#> 1: 1 0.1353093 33 0.6 0.6 4 0.2 18 -1
#> 2: 2 0.1282925 59 0.8 1.0 10 0.2 6 -1
#> 3: 3 0.2360723 100 0.8 0.8 4 0.1 2 -1
#> 4: 4 0.1298904 71 1.0 0.8 4 0.1 10 -1
#> 5: 5 0.1357692 32 1.0 0.6 6 0.2 18 -1
#> 6: 6 0.1313455 64 1.0 1.0 8 0.1 14 -1
#> verbose objective metric num_class
#> 1: -1 multiclass multi_logloss 3
#> 2: -1 multiclass multi_logloss 3
#> 3: -1 multiclass multi_logloss 3
#> 4: -1 multiclass multi_logloss 3
#> 5: -1 multiclass multi_logloss 3
#> 6: -1 multiclass multi_logloss 3
<- mlexperiments::MLTuneParameters$new(
tuner learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),strategy = "bayesian",
ncores = ncores,
seed = seed
)
$parameter_grid <- parameter_grid
tuner$parameter_bounds <- parameter_bounds
tuner
$learner_args <- learner_args
tuner$optim_args <- optim_args
tuner
$split_type <- "stratified"
tuner
$set_data(
tunerx = train_x,
y = train_y
)
<- tuner$execute(k = 3)
tuner_results_bayesian #>
#> Registering parallel backend using 4 cores.
head(tuner_results_bayesian)
#> Epoch setting_id bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves gpUtility acqOptimum inBounds
#> 1: 0 1 0.6 0.6 4 0.2 18 NA FALSE TRUE
#> 2: 0 2 0.8 1.0 10 0.2 6 NA FALSE TRUE
#> 3: 0 3 0.8 0.8 4 0.1 2 NA FALSE TRUE
#> 4: 0 4 1.0 0.8 4 0.1 10 NA FALSE TRUE
#> 5: 0 5 1.0 0.6 6 0.2 18 NA FALSE TRUE
#> 6: 0 6 1.0 1.0 8 0.1 14 NA FALSE TRUE
#> Elapsed Score metric_optim_mean nrounds errorMessage max_depth verbose objective metric num_class
#> 1: 1.283 -0.1353093 0.1353093 33 NA -1 -1 multiclass multi_logloss 3
#> 2: 1.300 -0.1282925 0.1282925 59 NA -1 -1 multiclass multi_logloss 3
#> 3: 1.277 -0.2360723 0.2360723 100 NA -1 -1 multiclass multi_logloss 3
#> 4: 1.460 -0.1298904 0.1298904 71 NA -1 -1 multiclass multi_logloss 3
#> 5: 0.360 -0.1357692 0.1357692 32 NA -1 -1 multiclass multi_logloss 3
#> 6: 0.561 -0.1313455 0.1313455 64 NA -1 -1 multiclass multi_logloss 3
<- mlexperiments::MLCrossValidation$new(
validator learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),fold_list = fold_list,
ncores = ncores,
seed = seed
)
$learner_args <- tuner$results$best.setting[-1]
validator
$predict_args <- predict_args
validator$performance_metric <- performance_metric
validator$performance_metric_args <- performance_metric_args
validator$return_models <- return_models
validator
$set_data(
validatorx = train_x,
y = train_y
)
<- validator$execute()
validator_results #>
#> CV fold: Fold1
#>
#> CV fold: Fold2
#> CV progress [====================================================================>-----------------------------------] 2/3 ( 67%)
#>
#> CV fold: Fold3
#> CV progress [========================================================================================================] 3/3 (100%)
#>
head(validator_results)
#> fold performance bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves nrounds max_depth verbose
#> 1: Fold1 0.9674260 0.8 0.6 8 0.1 14 66 -1 -1
#> 2: Fold2 0.9534347 0.8 0.6 8 0.1 14 66 -1 -1
#> 3: Fold3 0.9549840 0.8 0.6 8 0.1 14 66 -1 -1
#> objective metric num_class
#> 1: multiclass multi_logloss 3
#> 2: multiclass multi_logloss 3
#> 3: multiclass multi_logloss 3
<- mlexperiments::MLNestedCV$new(
validator learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),strategy = "grid",
fold_list = fold_list,
k_tuning = 3L,
ncores = ncores,
seed = seed
)
$parameter_grid <- parameter_grid
validator$learner_args <- learner_args
validator$split_type <- "stratified"
validator
$predict_args <- predict_args
validator$performance_metric <- performance_metric
validator$performance_metric_args <- performance_metric_args
validator$return_models <- return_models
validator
$set_data(
validatorx = train_x,
y = train_y
)
<- validator$execute()
validator_results #>
#> CV fold: Fold1
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#>
#> Parameter settings [============================>-------------------------------------------------------------------] 3/10 ( 30%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=====================================>----------------------------------------------------------] 4/10 ( 40%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [===============================================>------------------------------------------------] 5/10 ( 50%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=========================================================>--------------------------------------] 6/10 ( 60%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [==================================================================>-----------------------------] 7/10 ( 70%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [============================================================================>-------------------] 8/10 ( 80%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=====================================================================================>----------] 9/10 ( 90%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [===============================================================================================] 10/10 (100%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> CV fold: Fold2
#> CV progress [====================================================================>-----------------------------------] 2/3 ( 67%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [============================>-------------------------------------------------------------------] 3/10 ( 30%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=====================================>----------------------------------------------------------] 4/10 ( 40%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [===============================================>------------------------------------------------] 5/10 ( 50%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=========================================================>--------------------------------------] 6/10 ( 60%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [==================================================================>-----------------------------] 7/10 ( 70%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [============================================================================>-------------------] 8/10 ( 80%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=====================================================================================>----------] 9/10 ( 90%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [===============================================================================================] 10/10 (100%)
#> [LightGBM] [Info] Start training from score -1.423260
#> [LightGBM] [Info] Start training from score -1.427452
#> [LightGBM] [Info] Start training from score -0.655556
#> [LightGBM] [Info] Start training from score -1.428460
#> [LightGBM] [Info] Start training from score -1.420092
#> [LightGBM] [Info] Start training from score -0.656564
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> CV fold: Fold3
#> CV progress [========================================================================================================] 3/3 (100%)
#>
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [============================>-------------------------------------------------------------------] 3/10 ( 30%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=====================================>----------------------------------------------------------] 4/10 ( 40%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [===============================================>------------------------------------------------] 5/10 ( 50%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=========================================================>--------------------------------------] 6/10 ( 60%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [==================================================================>-----------------------------] 7/10 ( 70%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [============================================================================>-------------------] 8/10 ( 80%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [=====================================================================================>----------] 9/10 ( 90%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
#>
#> Parameter settings [===============================================================================================] 10/10 (100%)
#> [LightGBM] [Info] Start training from score -1.421241
#> [LightGBM] [Info] Start training from score -1.429645
#> [LightGBM] [Info] Start training from score -0.655482
#> [LightGBM] [Info] Start training from score -1.424424
#> [LightGBM] [Info] Start training from score -1.428634
#> [LightGBM] [Info] Start training from score -0.654471
#> [LightGBM] [Info] Start training from score -1.422251
#> [LightGBM] [Info] Start training from score -1.426444
#> [LightGBM] [Info] Start training from score -0.656491
head(validator_results)
#> fold performance nrounds bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves max_depth verbose
#> 1: Fold1 0.9674260 62 0.8 0.6 8 0.1 14 -1 -1
#> 2: Fold2 0.9506435 64 0.8 1.0 10 0.2 6 -1 -1
#> 3: Fold3 0.9559827 100 0.6 0.6 8 0.1 6 -1 -1
#> objective metric num_class
#> 1: multiclass multi_logloss 3
#> 2: multiclass multi_logloss 3
<- mlexperiments::MLNestedCV$new(
validator learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),strategy = "bayesian",
fold_list = fold_list,
k_tuning = 3L,
ncores = ncores,
seed = seed
)
$parameter_grid <- parameter_grid
validator$learner_args <- learner_args
validator$split_type <- "stratified"
validator
$parameter_bounds <- parameter_bounds
validator$optim_args <- optim_args
validator
$predict_args <- predict_args
validator$performance_metric <- performance_metric
validator$performance_metric_args <- performance_metric_args
validator$return_models <- TRUE
validator
$set_data(
validatorx = train_x,
y = train_y
)
<- validator$execute()
validator_results #>
#> CV fold: Fold1
#>
#> Registering parallel backend using 4 cores.
#>
#> CV fold: Fold2
#> CV progress [====================================================================>-----------------------------------] 2/3 ( 67%)
#>
#> Registering parallel backend using 4 cores.
#>
#> CV fold: Fold3
#> CV progress [========================================================================================================] 3/3 (100%)
#>
#> Registering parallel backend using 4 cores.
head(validator_results)
#> fold performance bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves nrounds max_depth verbose
#> 1: Fold1 0.9727424 0.2000000 0.2559232 2 0.1992638 20 52 -1 -1
#> 2: Fold2 0.9494955 0.8293692 0.5664407 7 0.1203691 11 76 -1 -1
#> 3: Fold3 0.9568462 0.4438889 0.3041453 10 0.1462295 11 67 -1 -1
#> objective metric num_class
#> 1: multiclass multi_logloss 3
#> 2: multiclass multi_logloss 3
#> 3: multiclass multi_logloss 3
<- mlexperiments::predictions(
preds_lightgbm object = validator,
newdata = test_x
)
<- mlexperiments::performance(
perf_lightgbm object = validator,
prediction_results = preds_lightgbm,
y_ground_truth = test_y
)
perf_lightgbm#> model performance
#> 1: Fold1 0.9596127
#> 2: Fold2 0.9612778
#> 3: Fold3 0.9583793
Here, lightgbm
’s weight
-argument is used to rescale the case-weights during the training.
# define the target weights
<- ifelse(train_y == 1, 0.8, ifelse(train_y == 2, 1.2, 1))
y_weights head(y_weights)
#> [1] 1.2 1.2 0.0 0.8 0.8 0.0
<- mlexperiments::MLTuneParameters$new(
tuner_w_weights learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),strategy = "grid",
ncores = ncores,
seed = seed
)
$parameter_grid <- parameter_grid
tuner_w_weights$learner_args <- c(
tuner_w_weights
learner_args,list(case_weights = y_weights)
)$split_type <- "stratified"
tuner_w_weights
$set_data(
tuner_w_weightsx = train_x,
y = train_y
)
<- tuner_w_weights$execute(k = 3)
tuner_results_grid #>
#> Parameter settings [============================>-------------------------------------------------------------------] 3/10 ( 30%)
#> Parameter settings [=====================================>----------------------------------------------------------] 4/10 ( 40%)
#> Parameter settings [===============================================>------------------------------------------------] 5/10 ( 50%)
#> Parameter settings [=========================================================>--------------------------------------] 6/10 ( 60%)
#> Parameter settings [==================================================================>-----------------------------] 7/10 ( 70%)
#> Parameter settings [============================================================================>-------------------] 8/10 ( 80%)
#> Parameter settings [=====================================================================================>----------] 9/10 ( 90%)
#> Parameter settings [===============================================================================================] 10/10 (100%)
head(tuner_results_grid)
#> setting_id metric_optim_mean nrounds bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves max_depth
#> <int> <num> <int> <num> <num> <num> <num> <num> <int>
#> 1: 1 0.1294454 29 0.6 0.6 4 0.2 18 -1
#> 2: 2 0.1221349 51 0.8 1.0 10 0.2 6 -1
#> 3: 3 0.2240799 100 0.8 0.8 4 0.1 2 -1
#> 4: 4 0.1194221 75 1.0 0.8 4 0.1 10 -1
#> 5: 5 0.1281037 32 1.0 0.6 6 0.2 18 -1
#> 6: 6 0.1245721 60 1.0 1.0 8 0.1 14 -1
#> verbose objective metric num_class
#> <int> <char> <char> <char>
#> 1: -1 multiclass multi_logloss 3
#> 2: -1 multiclass multi_logloss 3
#> 3: -1 multiclass multi_logloss 3
#> 4: -1 multiclass multi_logloss 3
#> 5: -1 multiclass multi_logloss 3
#> 6: -1 multiclass multi_logloss 3
<- mlexperiments::MLCrossValidation$new(
validator learner = mllrnrs::LearnerLightgbm$new(
metric_optimization_higher_better = FALSE
),fold_list = fold_list,
ncores = ncores,
seed = seed
)
# append the optimized setting from above with the newly created weights
$learner_args <- c(
validator$results$best.setting[-1],
tunerlist("case_weights" = y_weights)
)
$predict_args <- predict_args
validator$performance_metric <- performance_metric
validator$performance_metric_args <- performance_metric_args
validator$return_models <- return_models
validator
$set_data(
validatorx = train_x,
y = train_y
)
<- validator$execute()
validator_results #>
#> CV fold: Fold1
#>
#> CV fold: Fold2
#>
#> CV fold: Fold3
#> CV progress [========================================================================================================] 3/3 (100%)
#>
head(validator_results)
#> fold performance bagging_fraction feature_fraction min_data_in_leaf learning_rate num_leaves nrounds max_depth verbose
#> <char> <num> <num> <num> <num> <num> <num> <int> <int> <int>
#> 1: Fold1 0.9635665 0.8 0.6 8 0.1 14 66 -1 -1
#> 2: Fold2 0.9551452 0.8 0.6 8 0.1 14 66 -1 -1
#> 3: Fold3 0.9521232 0.8 0.6 8 0.1 14 66 -1 -1
#> objective metric num_class
#> <char> <char> <char>
#> 1: multiclass multi_logloss 3
#> 2: multiclass multi_logloss 3
#> 3: multiclass multi_logloss 3