Combining rpart hyper tuning parameters with down sampling in MLR3

I am walking through great examples from the MLR3 package (mlr3gallery:imbalanced data examples), and I was hoping to see an example that combines hyper parameter tuning AND an imbalance correction.

From the link above, as description of what I am trying to achieve:

To keep runtime low, we define the search space only for the imbalacy correction method. However, one can also jointly tune the hyperparameter of the learner along with the imbalance correction method by extending the search space with the learner’s hyperparameters.

Here is an example that comes close - mlr3 PipeOps: Create branches with different data transformations and benchmark different learners within and between branches

So we can (mis)use missuse's great example from this as a walkthough:

#packages
library(paradox)
library(mlr3)
library(mlr3pipelines)
library(mlr3tuning)

#set up an rpart learner
learner <- lrn("classif.rpart", predict_type = "prob")
learner$param_set$values <- list(
  cp = 0,
  maxdepth = 21,
  minbucket = 12,
  minsplit = 24
)

#Create the tree graphs:

# graph 1, just imputehist
graph_nop <- po("imputehist") %>>%
  learner

# graph 2 : imputehist and undersample majority class (ratio relative to majority class)

graph_down <- po("imputehist") %>>%
  po("classbalancing", id = "undersample", adjust = "major", 
     reference = "major", shuffle = FALSE, ratio = 1/2) %>>%
  learner

# graph 3: impute hist and oversample minority class (ratio relative to minority class)

graph_up <- po("imputehist") %>>%
  po("classbalancing", id = "oversample", adjust = "minor", 
     reference = "minor", shuffle = FALSE, ratio = 2) %>>%
  learner

#Convert graphs to learners and set predict_type

graph_nop <-  GraphLearner$new(graph_nop)
graph_nop$predict_type <- "prob"

graph_down <- GraphLearner$new(graph_down)
graph_down$predict_type <- "prob"

graph_up <- GraphLearner$new(graph_up)
graph_up$predict_type <- "prob"

#define re-sampling and instantiate it so always the same split will be used:

hld <- rsmp("holdout")

set.seed(123)
hld$instantiate(tsk("sonar"))

#Benchmark

bmr <- benchmark(design = benchmark_grid(task = tsk("sonar"),
                                         learner = list(graph_nop,
                                                        graph_up,
                                                        graph_down),
                                         hld),
                 store_models = TRUE) #only needed if you want to inspect the models

#check result using different measures:

  bmr$aggregate(msr("classif.auc"))
  bmr$aggregate(msr("classif.ce"))

#This can be also performed within one pipeline with branching but one would need to define the paramset and use a tuner:

  graph2 <- 
  po("imputehist") %>>%
  po("branch", c("nop", "classbalancing_up", "classbalancing_down")) %>>%
  gunion(list(
    po("nop", id = "nop"),
    po("classbalancing", id = "classbalancing_up", ratio = 2, reference = 'major'),
    po("classbalancing", id = "classbalancing_down", ratio = 2, reference = 'minor') 
  )) %>>%
  po("unbranch") %>>%
  learner

graph2$plot()

#Note that the unbranch happens before the learner since one (always the same) learner is being used. Convert graph to learner and set predict_type

graph2 <- GraphLearner$new(graph2)
graph2$predict_type <- "prob"

#Define the param set. In this case just the different branch options.

ps <- ParamSet$new(
  list(
    ParamFct$new("branch.selection", levels = c("nop", "classbalancing_up", "classbalancing_down")),
  ))


#In general you would want to add also learner hyper parameters like cp and minsplit for rpart as well as the ratio of over/undersampling.

So how do we add the learner hyper parameters like cp and minsplit at this point?

#perhaps by adding them to the param list?
ps = ParamSet$new(list(
  ParamFct$new("branch.selection", levels = c("nop", "classbalancing_up", "classbalancing_down")),
  ParamDbl$new("cp", lower = 0.001, upper = 0.1),
  ParamInt$new("minsplit", lower = 1, upper = 10)
))

#Create a tuning instance and grid search with resolution 1 since no other parameters are tuned. The tuner will iterate through different pipeline branches as defined in the paramset.

instance <- TuningInstance$new(
  task = tsk("sonar"),
  learner = graph2,
  resampling = hld,
  measures = msr("classif.auc"),
  param_set = ps,
  terminator = term("none")
)


tuner <- tnr("grid_search", resolution = 1)
set.seed(321)
tuner$tune(instance)

But this results in:

Error in (function (xs)  : 
  Assertion on 'xs' failed: Parameter 'cp' not available..

I feel I may be missing a branch layer on how to combine these two things (the rpart hyper parameters / minsplit and cp; and the down/up sampling)? Thank you for any assistance.

Solution

As soon as you construct a piped learner the IDs of the underlaying params change, as they are added a prefix. You can always check the param_set of the learner. In your example it is graph2$param_set. There you will see that the params you are looking for are the following:

ps = ParamSet$new(list(
  ParamFct$new("branch.selection", levels = c("nop", "classbalancing_up", "classbalancing_down")),
  ParamDbl$new("classif.rpart.cp", lower = 0.001, upper = 0.1),
  ParamInt$new("classif.rpart.minsplit", lower = 1, upper = 10)
))