mlr-org · mb706 · Aug 12, 2025 · Aug 13, 2025 · Aug 13, 2025 · Aug 13, 2025
diff --git a/.lintr b/.lintr
@@ -5,6 +5,7 @@ linters: linters_with_defaults(
     object_name_linter = object_name_linter(c("snake_case", "CamelCase")), # only allow snake case and camel case object names
     cyclocomp_linter = NULL, # do not check function complexity
     commented_code_linter = NULL, # allow code in comments
-    line_length_linter = line_length_linter(120)
+    line_length_linter = line_length_linter(120),
+    indentation_linter = indentation_linter(indent = 2, hanging_indent_style = "never")
     )
 
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -48,6 +48,10 @@ Suggests:
     survival,
     testthat (>= 3.0.0),
     withr
+Remotes:
+    mlr-org/mlr3misc@common_baseclass,
+    mlr-org/mlr3pipelines@common_baseclass,
+    mlr-org/mlr3@common_baseclass
 Config/testthat/edition: 3
 Encoding: UTF-8
 NeedsCompilation: no
@@ -71,6 +75,7 @@ Collate:
     'FilterJMI.R'
     'FilterJMIM.R'
     'FilterKruskalTest.R'
+    'FilterLearnerPerformance.R'
     'FilterMIM.R'
     'FilterMRMR.R'
     'FilterNJMIM.R'

diff --git a/R/Filter.R b/R/Filter.R
@@ -16,17 +16,8 @@
 #' @family Filter
 #' @export
 Filter = R6Class("Filter",
+  inherit = Mlr3Component,
   public = list(
-    #' @field id (`character(1)`)\cr
-    #'   Identifier of the object.
-    #'   Used in tables, plot and text output.
-    id = NULL,
-
-    #' @field label (`character(1)`)\cr
-    #'   Label for this object.
-    #'   Can be used in tables, plot and text output instead of the ID.
-    label = NA_character_,
-
     #' @field task_types (`character()`)\cr
     #'   Set of supported task types, e.g. `"classif"` or `"regr"`.
     #'   Can be set to the scalar value `NA` to allow any task type.
@@ -39,23 +30,10 @@ Filter = R6Class("Filter",
     #'   [mlr3::Task]task properties.
     task_properties = NULL,
 
-    #' @field param_set ([paradox::ParamSet])\cr
-    #'   Set of hyperparameters.
-    param_set = NULL,
-
     #' @field feature_types (`character()`)\cr
     #'   Feature types of the filter.
     feature_types = NULL,
 
-    #' @field packages ([character()])\cr
-    #'   Packages which this filter is relying on.
-    packages = NULL,
-
-    #' @field man (`character(1)`)\cr
-    #'   String in the format `[pkg]::[topic]` pointing to a manual page for this object.
-    #'   Defaults to `NA`, but can be set by child classes.
-    man = NULL,
-
     #' @field scores
     #'   Stores the calculated filter score values as named numeric vector.
     #'   The vector is sorted in decreasing order with possible `NA` values
@@ -90,31 +68,27 @@ Filter = R6Class("Filter",
     #'   String in the format `[pkg]::[topic]` pointing to a manual page for
     #'   this object. The referenced help package can be opened via method
     #'   `$help()`.
-    initialize = function(id, task_types, task_properties = character(),
-      param_set = ps(), feature_types = character(), packages = character(), label = NA_character_,
-      man = NA_character_) {
+    initialize = function(id = dict_entry, task_types, task_properties = character(0),
+      param_set = ps(), feature_types = character(0), packages = character(0),
+      properties = character(0), label, man, dict_entry = id
+    ) {
+      if (!missing(label) || !missing(man)) {
+        mlr3component_deprecation_msg("label and man are deprecated for Filter construction and will be removed in the future.")
+      }
+
+      super$initialize(dict_entry = dict_entry, dict_shortaccess = "flt", id = id,
+        param_set = param_set, packages = packages, properties = properties)
 
-      self$id = assert_string(id)
-      self$label = assert_string(label, na.ok = TRUE)
       if (!test_scalar_na(task_types)) {
         # we allow any task type here, otherwise we are not able to construct
         # the filter without loading additional packages like mlr3proba
         assert_character(task_types, any.missing = FALSE)
       }
+      assert_subset(properties, mlr_reflections$filter_properties)  # only allow missings for now
       self$task_types = task_types
       self$task_properties = assert_subset(task_properties, unlist(mlr_reflections$task_properties, use.names = FALSE))
-      self$param_set = assert_param_set(param_set)
       self$feature_types = assert_subset(feature_types, mlr_reflections$task_feature_types)
-      self$packages = assert_character(packages, any.missing = FALSE, min.chars = 1L)
       self$scores = set_names(numeric(), character())
-      self$man = assert_string(man, na.ok = TRUE)
-    },
-
-    #' @description
-    #' Format helper for Filter class
-    #' @param ... (ignored).
-    format = function(...) {
-      sprintf("<%s:%s>", class(self)[1L], self$id) # nocov
     },
 
     #' @description
@@ -133,12 +107,6 @@ Filter = R6Class("Filter",
       }
     },
 
-    #' @description
-    #' Opens the corresponding help page referenced by field `$man`.
-    help = function() {
-      open_help(self$man) # nocov
-    },
-
     #' @description
     #'   Calculates the filter score values for the provided [mlr3::Task] and
     #'   stores them in field `scores`. `nfeat` determines the minimum number of
@@ -200,39 +168,6 @@ Filter = R6Class("Filter",
 
       invisible(self)
     }
-  ),
-
-  active = list(
-    #' @field properties ([character()])\cr
-    #'   Properties of the filter. Currently, only `"missings"` is supported.
-    #'   A filter has the property `"missings"`, iff the filter can handle missing values
-    #'   in the features in a graceful way. Otherwise, an assertion is thrown if missing
-    #'   values are detected.
-    properties = function(rhs) {
-      assert_ro_binding(rhs)
-      get_properties = get0(".get_properties", private)
-      if (is.null(get_properties)) character() else get_properties()
-    },
-
-    #' @field hash (`character(1)`)\cr
-    #' Hash (unique identifier) for this object.
-    hash = function(rhs) {
-      assert_ro_binding(rhs)
-      calculate_hash(class(self), self$id, self$param_set$values, mget(private$.extra_hash, envir = self))
-    },
-
-    #' @field phash (`character(1)`)\cr
-    #' Hash (unique identifier) for this partial object, excluding some components
-    #' which are varied systematically during tuning (parameter values) or feature
-    #' selection (feature names).
-    phash = function(rhs) {
-      assert_ro_binding(rhs)
-      calculate_hash(class(self), self$id, mget(private$.extra_hash, envir = self))
-    }
-  ),
-
-  private = list(
-    .extra_hash = character()
   )
 )
 

diff --git a/R/FilterAUC.R b/R/FilterAUC.R
@@ -43,13 +43,10 @@ FilterAUC = R6Class("FilterAUC",
     #' @description Create a FilterAUC object.
     initialize = function() {
       super$initialize(
-        id = "auc",
+        dict_entry = "auc",
         task_types = "classif",
         task_properties = "twoclass",
-        feature_types = c("integer", "numeric"),
-        packages = "mlr3measures",
-        label = "Area Under the ROC Curve Score",
-        man = "mlr3filters::mlr_filters_auc"
+        feature_types = c("integer", "numeric")
       )
     }
   ),

diff --git a/R/FilterAnova.R b/R/FilterAnova.R
@@ -47,12 +47,10 @@ FilterAnova = R6Class("FilterAnova",
     #' @description Create a FilterAnova object.
     initialize = function() {
       super$initialize(
-        id = "anova",
+        dict_entry = "anova",
         packages = "stats",
         feature_types = c("integer", "numeric"),
-        task_types = "classif",
-        label = "ANOVA F-Test",
-        man = "mlr3filters::mlr_filters_anova"
+        task_types = "classif"
       )
     }
   ),

diff --git a/R/FilterBoruta.R b/R/FilterBoruta.R
@@ -49,20 +49,16 @@ FilterBoruta = R6Class("FilterBoruta",
         doTrace = p_int(lower = 0, upper = 4, default = 0),
         holdHistory = p_lgl(default = TRUE),
         getImp = p_uty(),
-        keep = p_fct(levels = c("confirmed", "tentative"), default = "confirmed"),
-        num.threads = p_int(lower = 1, default = 1)
+        keep = p_fct(levels = c("confirmed", "tentative"), tags = "required", init = "confirmed"),
+        num.threads = p_int(lower = 1, default = 2, init = 1)
       )
 
-      param_set$set_values(keep = "confirmed", num.threads = 1)
-
       super$initialize(
-        id = "boruta",
+        dict_entry = "boruta",
         task_types = c("regr", "classif"),
         param_set = param_set,
         packages = "Boruta",
-        feature_types = c("integer", "numeric"),
-        label = "Burota",
-        man = "mlr3filters::mlr_filters_boruta"
+        feature_types = c("integer", "numeric")
       )
     }
   ),

diff --git a/R/FilterCMIM.R b/R/FilterCMIM.R
@@ -46,18 +46,15 @@ FilterCMIM = R6Class("FilterCMIM",
     #' @description Create a FilterCMIM object.
     initialize = function() {
       param_set = ps(
-        threads = p_int(lower = 0L, default = 0L, tags = "threads")
+        threads = p_int(lower = 0L, default = 0L, tags = "threads", init = 1L)
       )
-      param_set$values = list(threads = 1L)
 
       super$initialize(
-        id = "cmim",
+        dict_entry = "cmim",
         task_types = c("classif", "regr"),
         param_set = param_set,
         feature_types = c("integer", "numeric", "factor", "ordered"),
-        packages = "praznik",
-        label = "Minimal Conditional Mutual Information Maximization",
-        man = "mlr3filters::mlr_filters_cmim"
+        packages = "praznik"
       )
     }
   ),

diff --git a/R/FilterCarScore.R b/R/FilterCarScore.R
@@ -53,13 +53,11 @@ FilterCarScore = R6Class("FilterCarScore",
       param_set$values = list(verbose = FALSE)
 
       super$initialize(
-        id = "carscore",
+        dict_entry = "carscore",
         task_types = "regr",
         param_set = param_set,
         feature_types = c("logical", "integer", "numeric"),
-        packages = "care",
-        label = "Correlation-Adjusted coRrelation Score",
-        man = "mlr3filters::mlr_filters_carscore"
+        packages = "care"
       )
     }
   ),

diff --git a/R/FilterCarSurvScore.R b/R/FilterCarSurvScore.R
@@ -24,13 +24,11 @@ FilterCarSurvScore = R6Class("FilterCarSurvScore",
         denom = p_fct(c("1/n", "sum_w"), default = "1/n")
       )
       super$initialize(
-        id = "surv.carsurvscore",
+        dict_entry = "carsurvscore",
         packages = c("carSurv", "mlr3proba"),
         param_set = ps,
         feature_types = c("integer", "numeric"),
-        task_types = "surv",
-        label = "Correlation-Adjusted coRrelation Survival Score",
-        man = "mlr3filters::mlr_filters_carsurvscore"
+        task_types = "surv"
       )
     }
   ),

diff --git a/R/FilterCorrelation.R b/R/FilterCorrelation.R
@@ -60,13 +60,12 @@ FilterCorrelation = R6Class("FilterCorrelation",
       )
 
       super$initialize(
-        id = "correlation",
+        dict_entry = "correlation",
         task_types = "regr",
         param_set = param_set,
         feature_types = c("integer", "numeric"),
         packages = "stats",
-        label = "Correlation",
-        man = "mlr3filters::mlr_filters_correlation"
+        properties = "missings"
       )
     }
   ),
@@ -80,12 +79,7 @@ FilterCorrelation = R6Class("FilterCorrelation",
         y = as.matrix(task$truth()),
         .args = pv)[, 1L]
       set_names(abs(score), fn)
-    },
-
-    .get_properties = function() {
-      "missings"
     }
-
   )
 )
 

diff --git a/R/FilterDISR.R b/R/FilterDISR.R
@@ -46,18 +46,15 @@ FilterDISR = R6Class("FilterDISR",
     #' @description Create a FilterDISR object.
     initialize = function() {
       param_set = ps(
-        threads = p_int(lower = 0L, default = 0L, tags = "threads")
+        threads = p_int(lower = 0L, default = 0L, tags = "threads", init = 1L)
       )
-      param_set$values = list(threads = 1L)
 
       super$initialize(
-        id = "disr",
+        dict_entry = "disr",
         task_types = c("classif", "regr"),
         param_set = param_set,
         feature_types = c("integer", "numeric", "factor", "ordered"),
-        packages = "praznik",
-        label = "Double Input Symmetrical Relevance",
-        man = "mlr3filters::mlr_filters_disr"
+        packages = "praznik"
       )
     }
   ),

diff --git a/R/FilterFindCorrelation.R b/R/FilterFindCorrelation.R
@@ -55,13 +55,12 @@ FilterFindCorrelation = R6Class("FilterFindCorrelation",
       )
 
       super$initialize(
-        id = "find_correlation",
+        dict_entry = "find_correlation",
         task_types = NA_character_,
         param_set = param_set,
         feature_types = c("integer", "numeric"),
         packages = "stats",
-        label = "Correlation-based Score",
-        man = "mlr3filters::mlr_filters_find_correlation"
+        properties = "missings"
       )
     }
   ),
@@ -94,14 +93,6 @@ FilterFindCorrelation = R6Class("FilterFindCorrelation",
       # The following has the correct names and values, BUT we need scores in
       # reverse order. Shift by 1 to get positive values.
       1 - apply(cm, 2, max)
-    },
-    .get_properties = function() {
-      use = self$param_set$values$use %??% "everything"
-      if (use %in% c("complete.obs", "pairwise.complete.obs")) {
-        "missings"
-      } else {
-        character(0)
-      }
     }
   )
 )