chore: release

.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()
     }
   )
 )

README.Rmd

@@ -15,7 +15,7 @@ Learners which support the extraction feature importance scores can be combined
 
 <!-- badges: start -->
 [![r-cmd-check](https://github.com/mlr-org/mlr3filters/actions/workflows/r-cmd-check.yml/badge.svg)](https://github.com/mlr-org/mlr3filters/actions/workflows/r-cmd-check.yml)
-[![CodeFactor](https://www.codefactor.io/repository/github/mlr-org/mlr3filters/badge)](https://www.codefactor.io/repository/github/mlr-org/mlr3filters)
+[![CRAN Status](https://www.r-pkg.org/badges/version-ago/mlr3filters)](https://cran.r-project.org/package=mlr3filters)
 [![StackOverflow](https://img.shields.io/badge/stackoverflow-mlr3-orange.svg)](https://stackoverflow.com/questions/tagged/mlr3)
 [![Mattermost](https://img.shields.io/badge/chat-mattermost-orange.svg)](https://lmmisld-lmu-stats-slds.srv.mwn.de/mlr_invite/)
 <!-- badges: end -->
@@ -45,7 +45,7 @@ library("mlr3filters")
 
 task = tsk("sonar")
 filter = flt("auc")
-as.data.table(filter$calculate(task))
+head(as.data.table(filter$calculate(task)))
 ```
 
 ### Implemented Filters

README.md

@@ -18,7 +18,8 @@ from this package for embedded feature selection.
 <!-- badges: start -->
 
 [![r-cmd-check](https://github.com/mlr-org/mlr3filters/actions/workflows/r-cmd-check.yml/badge.svg)](https://github.com/mlr-org/mlr3filters/actions/workflows/r-cmd-check.yml)
-[![CodeFactor](https://www.codefactor.io/repository/github/mlr-org/mlr3filters/badge)](https://www.codefactor.io/repository/github/mlr-org/mlr3filters)
+[![CRAN
+Status](https://www.r-pkg.org/badges/version-ago/mlr3filters)](https://cran.r-project.org/package=mlr3filters)
 [![StackOverflow](https://img.shields.io/badge/stackoverflow-mlr3-orange.svg)](https://stackoverflow.com/questions/tagged/mlr3)
 [![Mattermost](https://img.shields.io/badge/chat-mattermost-orange.svg)](https://lmmisld-lmu-stats-slds.srv.mwn.de/mlr_invite/)
 <!-- badges: end -->
@@ -48,71 +49,16 @@ library("mlr3filters")
 
 task = tsk("sonar")
 filter = flt("auc")
-as.data.table(filter$calculate(task))
+head(as.data.table(filter$calculate(task)))
 ```
 
-    ##     feature       score
-    ##  1:     V11 0.281136807
-    ##  2:     V12 0.242918176
-    ##  3:     V10 0.232701774
-    ##  4:     V49 0.231262190
-    ##  5:      V9 0.230844246
-    ##  6:     V48 0.206278443
-    ##  7:     V13 0.204699545
-    ##  8:     V51 0.199359153
-    ##  9:     V47 0.197408749
-    ## 10:     V52 0.188028234
-    ## 11:     V46 0.187053032
-    ## 12:     V45 0.172703631
-    ## 13:      V4 0.165227083
-    ## 14:     V36 0.165180645
-    ## 15:      V5 0.153710411
-    ## 16:      V1 0.152317266
-    ## 17:     V44 0.151063435
-    ## 18:     V21 0.144469211
-    ## 19:     V35 0.141450729
-    ## 20:      V8 0.141032785
-    ## 21:     V43 0.140893471
-    ## 22:     V37 0.128076530
-    ## 23:      V6 0.124640104
-    ## 24:     V20 0.123572026
-    ## 25:      V2 0.122782576
-    ## 26:     V50 0.121807374
-    ## 27:      V3 0.116513421
-    ## 28:     V14 0.114748769
-    ## 29:     V22 0.113680691
-    ## 30:     V58 0.098402526
-    ## 31:     V42 0.091436798
-    ## 32:     V34 0.090275843
-    ## 33:     V23 0.085632024
-    ## 34:      V7 0.085074765
-    ## 35:     V28 0.078944924
-    ## 36:     V31 0.078341228
-    ## 37:     V19 0.077366026
-    ## 38:     V53 0.071607690
-    ## 39:     V54 0.068031949
-    ## 40:     V56 0.060648277
-    ## 41:     V39 0.058094177
-    ## 42:     V33 0.056793907
-    ## 43:     V24 0.055447200
-    ## 44:     V59 0.054611312
-    ## 45:     V27 0.053543234
-    ## 46:     V15 0.053264605
-    ## 47:     V32 0.043187517
-    ## 48:     V16 0.031485093
-    ## 49:     V29 0.028605926
-    ## 50:     V17 0.026516207
-    ## 51:     V41 0.024472927
-    ## 52:     V18 0.018621714
-    ## 53:     V26 0.017507198
-    ## 54:     V38 0.014813783
-    ## 55:     V25 0.013838581
-    ## 56:     V60 0.012213244
-    ## 57:     V30 0.006965729
-    ## 58:     V57 0.004876010
-    ## 59:     V55 0.003204235
-    ## 60:     V40 0.002182595
-    ##     feature       score
+    ##    feature     score
+    ## 1:     V11 0.2811368
+    ## 2:     V12 0.2429182
+    ## 3:     V10 0.2327018
+    ## 4:     V49 0.2312622
+    ## 5:      V9 0.2308442
+    ## 6:     V48 0.2062784
 
 ### Implemented Filters