chore: release

Author: mllg

.editorconfig

@@ -2,7 +2,7 @@
 root = true
 
 [*]
-charset = utf8
+charset = utf-8
 end_of_line = lf
 insert_final_newline = true
 indent_style = space

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: mlr3filters
 Title: Filter Based Feature Selection for 'mlr3'
-Version: 0.6.0.9000
+Version: 0.7.0
 Authors@R: c(
     person("Patrick", "Schratz", , "patrick.schratz@gmail.com", role = "aut",
            comment = c(ORCID = "0000-0003-0748-6624")),
@@ -43,7 +43,6 @@ Suggests:
     testthat (>= 3.0.0),
     withr
 Config/testthat/edition: 3
-Config/testthat/parallel: true
 Encoding: UTF-8
 NeedsCompilation: no
 Roxygen: list(markdown = TRUE, r6 = TRUE)

NEWS.md

@@ -1,10 +1,7 @@
-<!-- NEWS.md is maintained by https://cynkra.github.io/fledge, do not edit -->
+# mlr3filters 0.7.0
 
-# mlr3filters 0.6.0.9000
-
-## Chore
-
-- do not use deprecated function from mlr3misc
+- Features are now checked for missing values to improve error messages (#140)
+- Removed deprecated functions
 - Use featureless learner in defaults (#124)
 
 

R/Filter.R

@@ -91,8 +91,8 @@ Filter = R6Class("Filter",
     #'   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_) {
+      param_set = ps(), feature_types = character(), packages = character(), label = NA_character_,
+      man = NA_character_) {
 
       self$id = assert_string(id)
       self$label = assert_string(label, na.ok = TRUE)
@@ -102,13 +102,9 @@ Filter = R6Class("Filter",
         assert_character(task_types, any.missing = FALSE)
       }
       self$task_types = task_types
-      self$task_properties = assert_subset(
-        task_properties,
-        unlist(mlr_reflections$task_properties, use.names = FALSE))
+      self$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$feature_types = assert_subset(feature_types, mlr_reflections$task_feature_types)
       self$packages = assert_character(packages, any.missing = FALSE, min.chars = 1L)
       self$scores = set_names(numeric(), character())
       self$man = assert_string(man, na.ok = TRUE)
@@ -126,6 +122,7 @@ Filter = R6Class("Filter",
     print = function() {
       catn(format(self), if (is.na(self$label)) "" else paste0(": ", self$label))
       catn(str_indent("Task Types:", self$task_types))
+      catn(str_indent("Properties:", self$properties))
       catn(str_indent("Task Properties:", self$task_properties))
       catn(str_indent("Packages:", self$packages))
       catn(str_indent("Feature types:", self$feature_types))
@@ -162,13 +159,15 @@ Filter = R6Class("Filter",
     calculate = function(task, nfeat = NULL) {
       task = assert_task(as_task(task),
         feature_types = self$feature_types,
-        task_properties = self$task_properties)
+        task_properties = self$task_properties
+      )
+
+      fn = task$feature_names
 
       if (!is_scalar_na(self$task_types) && task$task_type %nin% self$task_types) {
         stopf("Filter '%s' does not support the type '%s' of task '%s'",
           self$id, task$task_type, task$id)
       }
-      fn = task$feature_names
 
       if (task$nrow == 0L) {
         self$scores = shuffle(set_names(rep.int(NA_real_, length(fn)), fn))
@@ -182,7 +181,7 @@ Filter = R6Class("Filter",
           nfeat = min(nfeat, length(fn))
         }
 
-        if (any(task$missings() > 0L)) {
+        if ("missings" %nin% self$properties && any(task$missings() > 0L)) {
           stopf("Cannot apply filter '%s' on task '%s', missing values detected",
             self$id, task$id)
         }
@@ -201,6 +200,19 @@ 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()
+    }
   )
 )
 

R/FilterImportance.R

@@ -22,7 +22,7 @@
 #' if (mlr3misc::require_namespaces(c("mlr3pipelines", "rpart", "mlr3learners"), quietly = TRUE)) {
 #'   library("mlr3learners")
 #'   library("mlr3pipelines")
-#'   task = mlr3::tsk("spam")
+#'   task = mlr3::tsk("sonar")
 #'
 #'   learner = mlr3::lrn("classif.rpart")
 #'
@@ -66,6 +66,10 @@ FilterImportance = R6Class("FilterImportance",
       learner = self$learner$clone(deep = TRUE)
       learner = learner$train(task = task)
       learner$base_learner()$importance()
+    },
+
+    .get_properties = function() {
+      intersect("missings", self$learner$properties)
     }
   )
 )

R/FilterPerformance.R

@@ -19,9 +19,10 @@
 #'   filter$calculate(task)
 #'   as.data.table(filter)
 #' }
+#'
 #' if (mlr3misc::require_namespaces(c("mlr3pipelines", "rpart"), quietly = TRUE)) {
 #'   library("mlr3pipelines")
-#'   task = mlr3::tsk("spam")
+#'   task = mlr3::tsk("iris")
 #'   l = lrn("classif.rpart")
 #'
 #'   # Note: `filter.frac` is selected randomly and should be tuned.
@@ -86,6 +87,10 @@ FilterPerformance = R6Class("FilterPerformance",
       }
 
       set_names(perf, fn)
+    },
+
+    .get_properties = function() {
+      intersect("missings", self$learner$properties)
     }
   )
 )

R/FilterPermutation.R

@@ -129,8 +129,13 @@ FilterPermutation = R6Class("FilterPermutation",
       }
 
       delta
+    },
+
+    .get_properties = function() {
+      intersect("missings", self$learner$properties)
     }
   )
+
 )
 
 #' @include mlr_filters.R

R/FilterRelief.R

@@ -11,7 +11,7 @@
 #' @examples
 #' if (requireNamespace("FSelectorRcpp")) {
 #'   ## Relief (default)
-#'   task = mlr3::tsk("sonar")
+#'   task = mlr3::tsk("iris")
 #'   filter = flt("relief")
 #'   filter$calculate(task)
 #'   head(filter$scores, 3)

R/FilterSelectedFeatures.R

@@ -29,7 +29,7 @@
 #' if (mlr3misc::require_namespaces(c("mlr3pipelines", "mlr3learners", "rpart"), quietly = TRUE)) {
 #'   library("mlr3pipelines")
 #'   library("mlr3learners")
-#'   task = mlr3::tsk("spam")
+#'   task = mlr3::tsk("sonar")
 #'
 #'   filter = flt("selected_features", learner = lrn("classif.rpart"))
 #'
@@ -75,6 +75,10 @@ FilterSelectedFeatures = R6Class("FilterSelectedFeatures",
       learner = learner$train(task = task)
       score = named_vector(task$feature_names, init = 0)
       replace(score, names(score) %in% learner$selected_features(), 1)
+    },
+
+    .get_properties = function() {
+      intersect("missings", self$learner$properties)
     }
   )
 )

R/FilterVariance.R

@@ -61,6 +61,10 @@ FilterVariance = R6Class("FilterVariance",
     .calculate = function(task, nfeat) {
       na_rm = self$param_set$values$na.rm %??% TRUE
       map_dbl(task$data(cols = task$feature_names), var, na.rm = na_rm)
+    },
+
+    .get_properties = function() {
+      if (isTRUE(self$param_set$values$na.rm)) "missings" else character()
     }
   )
 )