glmnet: Regression

library(mlexperiments)
library(mllrnrs)

See https://github.com/kapsner/mllrnrs/blob/main/R/learner_glmnet.R for implementation details.

Preprocessing

Import and Prepare Data

library(mlbench)
data("BostonHousing")
dataset <- BostonHousing |>
  data.table::as.data.table() |>
  na.omit()

feature_cols <- colnames(dataset)[1:13]
target_col <- "medv"

General Configurations

seed <- 123
if (isTRUE(as.logical(Sys.getenv("_R_CHECK_LIMIT_CORES_")))) {
  # on cran
  ncores <- 2L
} else {
  ncores <- ifelse(
    test = parallel::detectCores() > 4,
    yes = 4L,
    no = ifelse(
      test = parallel::detectCores() < 2L,
      yes = 1L,
      no = parallel::detectCores()
    )
  )
}
options("mlexperiments.bayesian.max_init" = 10L)

Generate Training- and Test Data

data_split <- splitTools::partition(
  y = dataset[, get(target_col)],
  p = c(train = 0.7, test = 0.3),
  type = "stratified",
  seed = seed
)

train_x <- model.matrix(
  ~ -1 + .,
  dataset[data_split$train, .SD, .SDcols = feature_cols]
)
train_y <- log(dataset[data_split$train, get(target_col)])


test_x <- model.matrix(
  ~ -1 + .,
  dataset[data_split$test, .SD, .SDcols = feature_cols]
)
test_y <- log(dataset[data_split$test, get(target_col)])

Generate Training Data Folds

fold_list <- splitTools::create_folds(
  y = train_y,
  k = 3,
  type = "stratified",
  seed = seed
)

Experiments

Prepare Experiments

# required learner arguments, not optimized
learner_args <- list(
  family = "gaussian",
  type.measure = "mse",
  standardize = TRUE
)

# set arguments for predict function and performance metric,
# required for mlexperiments::MLCrossValidation and
# mlexperiments::MLNestedCV
predict_args <- list(type = "response")
performance_metric <- metric("rmsle")
performance_metric_args <- NULL
return_models <- FALSE

# required for grid search and initialization of bayesian optimization
parameter_grid <- expand.grid(
  alpha = seq(0, 1, 0.05)
)
# reduce to a maximum of 10 rows
if (nrow(parameter_grid) > 10) {
  set.seed(123)
  sample_rows <- sample(seq_len(nrow(parameter_grid)), 10, FALSE)
  parameter_grid <- kdry::mlh_subset(parameter_grid, sample_rows)
}

# required for bayesian optimization
parameter_bounds <- list(
  alpha = c(0., 1.)
)
optim_args <- list(
  iters.n = ncores,
  kappa = 3.5,
  acq = "ucb"
)

Hyperparameter Tuning

Bayesian Optimization

tuner <- mlexperiments::MLTuneParameters$new(
  learner = mllrnrs::LearnerGlmnet$new(
    metric_optimization_higher_better = FALSE
  ),
  strategy = "bayesian",
  ncores = ncores,
  seed = seed
)

tuner$parameter_grid <- parameter_grid
tuner$parameter_bounds <- parameter_bounds

tuner$learner_args <- learner_args
tuner$optim_args <- optim_args

tuner$split_type <- "stratified"

tuner$set_data(
  x = train_x,
  y = train_y
)

tuner_results_bayesian <- tuner$execute(k = 3)
#> 
#> Registering parallel backend using 4 cores.

head(tuner_results_bayesian)
#>    Epoch setting_id alpha gpUtility acqOptimum inBounds Elapsed       Score metric_optim_mean       lambda errorMessage   family
#> 1:     0          1  0.70        NA      FALSE     TRUE   0.991 -0.03927487        0.03927487 0.0004916239           NA gaussian
#> 2:     0          2  0.90        NA      FALSE     TRUE   0.962 -0.03926677        0.03926677 0.0003174538           NA gaussian
#> 3:     0          3  0.65        NA      FALSE     TRUE   0.976 -0.03926382        0.03926382 0.0004005028           NA gaussian
#> 4:     0          4  0.10        NA      FALSE     TRUE   0.962 -0.03924418        0.03924418 0.0021612791           NA gaussian
#> 5:     0          5  0.45        NA      FALSE     TRUE   0.023 -0.03926592        0.03926592 0.0006968102           NA gaussian
#> 6:     0          6  0.05        NA      FALSE     TRUE   0.025 -0.03923310        0.03923310 0.0029793717           NA gaussian
#>    type.measure standardize
#> 1:          mse        TRUE
#> 2:          mse        TRUE
#> 3:          mse        TRUE
#> 4:          mse        TRUE
#> 5:          mse        TRUE
#> 6:          mse        TRUE

k-Fold Cross Validation

validator <- mlexperiments::MLCrossValidation$new(
  learner = mllrnrs::LearnerGlmnet$new(
    metric_optimization_higher_better = FALSE
  ),
  fold_list = fold_list,
  ncores = ncores,
  seed = seed
)

validator$learner_args <- tuner$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(
  x = train_x,
  y = train_y
)

validator_results <- validator$execute()
#> 
#> CV fold: Fold1
#> 
#> CV fold: Fold2
#> 
#> CV fold: Fold3

head(validator_results)
#>     fold performance      alpha      lambda   family type.measure standardize
#> 1: Fold1  0.05530167 0.01159355 0.004207556 gaussian          mse        TRUE
#> 2: Fold2  0.05239743 0.01159355 0.004207556 gaussian          mse        TRUE
#> 3: Fold3  0.05055533 0.01159355 0.004207556 gaussian          mse        TRUE

Nested Cross Validation

Inner Bayesian Optimization

validator <- mlexperiments::MLNestedCV$new(
  learner = mllrnrs::LearnerGlmnet$new(
    metric_optimization_higher_better = FALSE
  ),
  strategy = "bayesian",
  fold_list = fold_list,
  k_tuning = 3L,
  ncores = ncores,
  seed = 312
)

validator$parameter_grid <- parameter_grid
validator$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(
  x = train_x,
  y = train_y
)

validator_results <- validator$execute()
#> 
#> 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       alpha      lambda   family type.measure standardize
#> 1: Fold1  0.05541775 0.001528976 0.022251620 gaussian          mse        TRUE
#> 2: Fold2  0.05293442 0.001528976 0.022305296 gaussian          mse        TRUE
#> 3: Fold3  0.05056405 0.036876500 0.002985073 gaussian          mse        TRUE

Holdout Test Dataset Performance

Predict Outcome in Holdout Test Dataset

preds_glmnet <- mlexperiments::predictions(
  object = validator,
  newdata = test_x
)

Evaluate Performance on Holdout Test Dataset

perf_glmnet <- mlexperiments::performance(
  object = validator,
  prediction_results = preds_glmnet,
  y_ground_truth = test_y,
  type = "regression"
)
perf_glmnet
#>    model performance        mse        msle       mae       mape      rmse      rmsle       rsq      sse
#> 1: Fold1  0.05117877 0.03938447 0.002619267 0.1365514 0.04579938 0.1984552 0.05117877 0.7438377 6.104593
#> 2: Fold2  0.05218917 0.03992086 0.002723709 0.1407370 0.04763746 0.1998021 0.05218917 0.7403489 6.187734
#> 3: Fold3  0.04952504 0.03651949 0.002452730 0.1373768 0.04651953 0.1911007 0.04952504 0.7624719 5.660522