install.packages("mlr3hyperband")

remotes::install_github("mlr-org/mlr3hyperband")

library(mlr3hyperband)
library(mlr3learners)

# load learner, define search space and tag budget hyperparameter
learner = lrn("classif.xgboost",
  nrounds           = to_tune(p_int(27, 243, tags = "budget")),
  eta               = to_tune(1e-4, 1, logscale = TRUE),
  max_depth         = to_tune(1, 20),
  colsample_bytree  = to_tune(1e-1, 1),
  colsample_bylevel = to_tune(1e-1, 1),
  lambda            = to_tune(1e-3, 1e3, logscale = TRUE),
  alpha             = to_tune(1e-3, 1e3, logscale = TRUE),
  subsample         = to_tune(1e-1, 1)
)

# set parallel backend
future::plan("multisession")

# hyperparameter tuning on the pima indians diabetes data set
instance = tune(
  method = "hyperband",
  task = tsk("pima"),
  learner = learner,
  resampling = rsmp("cv", folds = 3),
  measures = msr("classif.ce"),
  eta = 3
)

library(mlr3hyperband)
library(mlr3learners)
library(mlr3pipelines)

# load learner and define search space
learner = lrn("classif.rpart",
  minsplit  = to_tune(2, 128, logscale = TRUE),
  minbucket = to_tune(1, 64, logscale = TRUE),
  cp        = to_tune(1e-04, 1e-1, logscale = TRUE)
)

# create graph learner with subsampling
graph_learner = as_learner(po("subsample") %>>% learner)

# increase subsample rate from ~ 3.4% to 100%
graph_learner$param_set$values$subsample.frac = to_tune(p_dbl(3^-3, 1, tags = "budget"))

# set parallel backend
future::plan("multisession")

# hyperparameter tuning on the spam data set
instance = tune(
  method = "hyperband",
  task = tsk("spam"),
  learner = graph_learner,
  resampling = rsmp("cv", folds = 3),
  measures = msr("classif.ce"),
  eta = 3
)

library(bbotk)
library(mlr3hyperband)

# define hyperparameter and budget parameter
search_space = domain = ps(
  x1        = p_dbl(-5, 10),
  x2        = p_dbl(0, 15),
  fidelity  = p_dbl(1e-2, 1, tags = "budget")
)

# modified branin function
objective = ObjectiveRFun$new(
  fun = branin,
  domain = domain,
  codomain = ps(y = p_dbl(tags = "minimize"))
)

# optimize branin function with hyperband
result = bb_optimize(objective, method = "hyperband", search_space = search_space,
  term_evals = NULL, eta = 2)

# optimized parameters
result$par

# optimal outcome
result$value