Tidy pkg, prepare for release (#54)

Author: pat-s

.Rbuildignore

@@ -11,3 +11,4 @@
 ^\.ignore$
 ^docs$
 ^pkgdown$
+.pre-commit-config.yaml

.github/CONTRIBUTING.md

@@ -2,7 +2,6 @@
 # History files
 .Rhistory
 .Rapp.history
-
 # Session Data files
 .RData
 # Example code in package build process
@@ -26,3 +25,4 @@ vignettes/*.pdf
 *.knit.md
 # Shiny token, see https://shiny.rstudio.com/articles/shinyapps.html
 rsconnect/
+.pre-commit-config.yaml

.github/ISSUE_TEMPLATE.md

@@ -17,8 +17,10 @@ Authors@R:
              role = "aut",
              email = "bernd_bischl@gmx.net",
              comment = c(ORCID = "0000-0001-6002-6980")))
-Description: Implements methods for feature selection via
-    filtering in 'mlr3'.
+Description: Extends mlr3 with filter methods for feature
+    selection.  Besides standalone filter methods (for example MRMR, CMIM)
+    built-in methods of any machine-learning algorithm are supported.
+    Partial scoring of multivariate filter methods is supported.
 License: LGPL-3
 URL: https://mlr3filters.mlr-org.com, https://github.com/mlr-org/mlr3filters
 BugReports: https://github.com/mlr-org/mlr3filters/issues

.gitignore

@@ -4,18 +4,17 @@
 #' @format [R6::R6Class] inheriting from [Filter].
 #' @include Filter.R
 #'
-#' @description
-#' Area under the (ROC) Curve filter calling [Metrics::auc()] from package \CRANpkg{Metrics}.
-#' Returns the absolute value of the difference between the AUC and 0.5.
+#' @description Area under the (ROC) Curve filter calling [Metrics::auc()] from
+#' package \CRANpkg{Metrics}. Returns the absolute value of the difference
+#' between the AUC and 0.5.
 #'
 #' @family Filter
 #' @export
 #' @examples
 #' task = mlr3::mlr_tasks$get("pima")
 #' filter = FilterAUC$new()
 #' filter$calculate(task)
-#' head(filter$scores, 3)
-#' as.data.table(filter)
+#' head(as.data.table(filter), 3)
 FilterAUC = R6Class("FilterAUC", inherit = Filter,
   public = list(
     initialize = function(id = "auc") {

DESCRIPTION

@@ -4,20 +4,19 @@
 #' @format [R6::R6Class] inheriting from [Filter].
 #' @include Filter.R
 #'
-#' @description
-#' ANOVA F-Test filter calling [stats::aov()].
-#' Note that this is equivalent to a t-test for binary classification.
+#' @description ANOVA F-Test filter calling [stats::aov()]. Note that this is
+#' equivalent to a t-test for binary classification.
 #'
-#' The filter value is `-log10(p)` where `p` is the p-value.
-#' This transformation is necessary to ensure numerical stability for very small p-values.
+#' The filter value is `-log10(p)` where `p` is the p-value. This transformation
+#' is necessary to ensure numerical stability for very small p-values.
 #'
 #' @family Filter
 #' @export
 #' @examples
 #' task = mlr3::mlr_tasks$get("iris")
 #' filter = FilterAnova$new()
 #' filter$calculate(task)
-#' as.data.table(filter)
+#' head(as.data.table(filter), 3)
 #'
 #' # transform to p-value
 #' 10^(-filter$scores)

R/FilterAUC.R

@@ -4,8 +4,8 @@
 #' @format [R6::R6Class] inheriting from [Filter].
 #' @include Filter.R
 #'
-#' @description
-#' Minimal conditional mutual information maximisation filter calling [praznik::CMIM()] from package \CRANpkg{praznik}.
+#' @description Minimal conditional mutual information maximisation filter
+#' calling [praznik::CMIM()] from package \CRANpkg{praznik}.
 #'
 #' This filter supports partial scoring (see [Filter]).
 #'
@@ -15,8 +15,6 @@
 #' task = mlr3::mlr_tasks$get("iris")
 #' filter = FilterCMIM$new()
 #' filter$calculate(task, nfeat = 2)
-#'
-#' head(filter$scores, 2)
 #' as.data.table(filter)
 FilterCMIM = R6Class("FilterCMIM", inherit = Filter,
   public = list(

R/FilterAnova.R

@@ -4,11 +4,11 @@
 #' @format [R6::R6Class] inheriting from [Filter].
 #' @include Filter.R
 #'
-#' @description
-#' Calculates the Correlation-Adjusted (marginal) coRelation scores (short CAR scores)
-#' implemented in [care::carscore()] in package \CRANpkg{care}.
-#' The CAR scores for a set of features are defined as the correlations between the target
-#' and the decorrelated features. The filter returns the absolute value of the calculated scores.
+#' @description Calculates the Correlation-Adjusted (marginal) coRelation scores
+#' (short CAR scores) implemented in [care::carscore()] in package
+#' \CRANpkg{care}. The CAR scores for a set of features are defined as the
+#' correlations between the target and the decorrelated features. The filter
+#' returns the absolute value of the calculated scores.
 #'
 #' Argument `verbose` defaults to `FALSE`.
 #'
@@ -18,8 +18,13 @@
 #' task = mlr3::mlr_tasks$get("mtcars")
 #' filter = FilterCarScore$new()
 #' filter$calculate(task)
-#' head(filter$scores, 3)
-#' as.data.table(filter)
+#' head(as.data.table(filter), 3)
+#'
+#' ## changing filter settings
+#' filter = FilterCarScore$new()
+#' filter$param_set$values = list("diagonal" = TRUE)
+#' filter$calculate(task)
+#' head(as.data.table(filter), 3)
 FilterCarScore = R6Class("FilterCarScore", inherit = Filter,
   public = list(
     initialize = function(id = "carscore") {
@@ -31,9 +36,9 @@ FilterCarScore = R6Class("FilterCarScore", inherit = Filter,
         param_set = ParamSet$new(list(
           ParamDbl$new("lambda", lower = 0, upper = 1, default = NO_DEF),
           ParamLgl$new("diagonal", default = FALSE),
-          ParamLgl$new("verbose", default = TRUE))),
-        param_vals = list(verbose = FALSE)
+          ParamLgl$new("verbose", default = TRUE)))
       )
+      self$param_set$values = list(verbose = FALSE)
     },
 
     calculate_internal = function(task, nfeat) {

R/FilterCMIM.R

@@ -11,10 +11,17 @@
 #' @family Filter
 #' @export
 #' @examples
+#' ## Pearson (default)
 #' task = mlr3::mlr_tasks$get("mtcars")
 #' filter = FilterCorrelation$new()
 #' filter$calculate(task)
 #' as.data.table(filter)
+#'
+#' ## Spearman
+#' filter = FilterCorrelation$new()
+#' filter$param_set$values = list("method" = "spearman")
+#' filter$calculate(task)
+#' as.data.table(filter)
 FilterCorrelation = R6Class("FilterCorrelation", inherit = Filter,
   public = list(
     initialize = function(id = "correlation") {