Merge branch 'main' into mauc

mlr-org · Aug 20, 2024 · ecf5f70 · ecf5f70
2 parents def0106 + 5fb059a
commit ecf5f70
Show file tree

Hide file tree

Showing 109 changed files with 841 additions and 290 deletions.
diff --git a/.Rbuildignore b/.Rbuildignore
@@ -11,3 +11,7 @@
 ^\.ccache$
 ^\.lintr$
 ^docs$
+^scratch$
+^revdep$
+^CRAN-SUBMISSION$
+^cran-comments\.md$
diff --git a/.gitignore b/.gitignore
@@ -1,5 +1,100 @@
-.Rproj.user
+# File created using '.gitignore Generator' for Visual Studio Code: https://bit.ly/vscode-gig
+# Created by https://www.toptal.com/developers/gitignore/api/visualstudiocode,linux,r
+# Edit at https://www.toptal.com/developers/gitignore?templates=visualstudiocode,linux,r
+
+### Linux ###
+*~
+
+# temporary files which can be created if a process still has a handle open of a deleted file
+.fuse_hidden*
+
+# KDE directory preferences
+.directory
+
+# Linux trash folder which might appear on any partition or disk
+.Trash-*
+
+# .nfs files are created when an open file is removed but is still being accessed
+.nfs*
+
+### R ###
+# History files
 .Rhistory
+.Rapp.history
+
+# Session Data files
 .RData
+.RDataTmp
+
+# User-specific files
 .Ruserdata
+
+# Example code in package build process
+*-Ex.R
+
+# Output files from R CMD build
+/*.tar.gz
+
+# Output files from R CMD check
+/*.Rcheck/
+
+# RStudio files
+.Rproj.user/
+
+# produced vignettes
+vignettes/*.html
+vignettes/*.pdf
+
+# OAuth2 token, see https://github.com/hadley/httr/releases/tag/v0.3
+.httr-oauth
+
+# knitr and R markdown default cache directories
+*_cache/
+/cache/
+
+# Temporary files created by R markdown
+*.utf8.md
+*.knit.md
+
+# R Environment Variables
+.Renviron
+
+# pkgdown site
 docs/
+
+# translation temp files
+po/*~
+
+# RStudio Connect folder
+rsconnect/
+
+### R.Bookdown Stack ###
+# R package: bookdown caching files
+/*_files/
+
+### VisualStudioCode ###
+.vscode/*
+!.vscode/settings.json
+!.vscode/tasks.json
+!.vscode/launch.json
+!.vscode/extensions.json
+!.vscode/*.code-snippets
+
+# Local History for Visual Studio Code
+.history/
+
+# Built Visual Studio Code Extensions
+*.vsix
+
+### VisualStudioCode Patch ###
+# Ignore all local history of files
+.history
+.ionide
+
+# End of https://www.toptal.com/developers/gitignore/api/visualstudiocode,linux,r
+
+# Custom rules (everything added below won't be overriden by 'Generate .gitignore File' if you use 'Update' option)
+
+revdep/
+cran-comments.md
+CRAN-SUBMISSION
diff --git a/.lintr b/.lintr
@@ -1,10 +1,10 @@
-linters: with_defaults(
-    # lintr defaults: https://github.com/jimhester/lintr#available-linters
+linters: linters_with_defaults(
+    # lintr defaults: https://lintr.r-lib.org/reference/default_linters.html
     # the following setup changes/removes certain linters
     assignment_linter = NULL, # do not force using <- for assignments
     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(120L)
     )
 
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,18 +1,16 @@
 Package: mlr3measures
 Title: Performance Measures for 'mlr3'
-Version: 0.5.0-9000
-Authors@R:
-    c(person(given = "Michel",
-             family = "Lang",
-             role = c("cre", "aut"),
-             email = "michellang@gmail.com",
-             comment = c(ORCID = "0000-0001-9754-0393")),
-      person(given = "Martin",
-             family = "Binder",
-             role = "ctb",
-             email = "mlr.developer@mb706.com"))
-Description: Implements multiple performance measures for
-    supervised learning.  Includes over 40 measures for regression and
+Version: 0.6.0.9000
+Authors@R: c(
+    person("Michel", "Lang", , "michellang@gmail.com", role = "aut",
+           comment = c(ORCID = "0000-0001-9754-0393")),
+    person("Martin", "Binder", , "mlr.developer@mb706.com", role = "ctb"),
+    person("Marc", "Becker", , "marcbecker@posteo.de", role = c("cre", "aut"),
+           comment = c(ORCID = "0000-0002-8115-0400")),
+    person("Lona", "Koers", , , role = "aut")
+  )
+Description: Implements multiple performance measures for supervised
+    learning.  Includes over 40 measures for regression and
     classification. Additionally, meta information about the performance
     measures can be queried, e.g. what the best and worst possible
     performances scores are.
@@ -28,10 +26,10 @@ Imports:
     PRROC
 Suggests:
     testthat (>= 3.0.0)
-Encoding: UTF-8
 Config/testthat/edition: 3
+Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.3.1
+RoxygenNote: 7.3.2
 Collate:
     'assertions.R'
     'bibentries.R'
@@ -48,7 +46,6 @@ Collate:
     'binary_fpr.R'
     'binary_gmean.R'
     'binary_gpr.R'
-    'binary_mcc.R'
     'binary_npv.R'
     'binary_ppv.R'
     'binary_prauc.R'
@@ -62,13 +59,15 @@ Collate:
     'classif_ce.R'
     'classif_logloss.R'
     'classif_mbrier.R'
+    'classif_mcc.R'
     'classif_zero_one.R'
     'confusion_matrix.R'
     'helper.R'
     'regr_ae.R'
     'regr_ape.R'
     'regr_bias.R'
     'regr_ktau.R'
+    'regr_linex.R'
     'regr_mae.R'
     'regr_mape.R'
     'regr_maxae.R'
@@ -78,6 +77,7 @@ Collate:
     'regr_mse.R'
     'regr_msle.R'
     'regr_pbias.R'
+    'regr_pinball.R'
     'regr_rae.R'
     'regr_rmse.R'
     'regr_rmsle.R'

diff --git a/NAMESPACE b/NAMESPACE
@@ -22,6 +22,7 @@ export(gmean)
 export(gpr)
 export(jaccard)
 export(ktau)
+export(linex)
 export(logloss)
 export(mae)
 export(mape)
@@ -40,8 +41,10 @@ export(medse)
 export(mse)
 export(msle)
 export(npv)
+export(one_zero)
 export(pbias)
 export(phi)
+export(pinball)
 export(ppv)
 export(prauc)
 export(precision)

diff --git a/NEWS.md b/NEWS.md
@@ -1,7 +1,13 @@
-# bbotk (development version)
+# mlr3measures (development version)
+
+* Added new measure `linex` (Linear-Exponential Loss).
+* Added new measure `pinball` (Average Pinball Loss).
+* Added new measure `mauc_mu` (Mu AUC).
+
+# mlr3measures 0.6.0
 
-* Added multiclass mu AUC measure `mauc_mu`.
 * Added binary classification measures `gmean` and `gpr`.
+* Added new measure `mcc` (Matthews Correlation Coefficient).
 
 # mlr3measures 0.5.0
 
@@ -13,15 +19,13 @@
 * Calculation of `fbeta` is now more numerically more stable.
 * Improved documentation.
 
-
 # mlr3measures 0.4.0
 
 * New measures to calculate set similarities: `jaccard` and `phi`.
 * Many measures now support sample weights.
   Supported measures have `$sample_weights` set to `TRUE` in the `measures`
   environment.
 
-
 # mlr3measures 0.3.1
 
 * Fix edge case for `msle` and `rmsle`.

diff --git a/R/bibentries.R b/R/bibentries.R
@@ -193,5 +193,15 @@ bibentries = c( # nolint start
     volume           = "97",
     series           = "Proceedings of Machine Learning Research",
     publisher        = "PMLR"
+  ),
+  varian_1975 = bibentry("incollection",
+    title        = "A Bayesian Approach to Real Estate Assessment",
+    author       = "Varian, Hal R.",
+    booktitle    = "Studies in Bayesian Econometrics and Statistics: In Honor of Leonard J. Savage",
+    editor       = "Stephen E. Fienberg and Arnold Zellner",
+    year         = "1975",
+    publisher    = "North-Holland",
+    address      = "Amsterdam",
+    pages        = "195--208"
   )
 ) # nolint end
diff --git a/R/binary_bbrier.R b/R/binary_bbrier.R
@@ -2,14 +2,12 @@
 #'
 #' @details
 #' The Binary Brier Score is defined as \deqn{
-#'    \frac{1}{n} \sum_{i=1}^n w_i (I_i - p_i)^2.
+#'    \frac{1}{n} \sum_{i=1}^n w_i (I_i - p_i)^2,
 #' }{
-#'    weighted.mean(((t == positive) - p)^2, w).
-#' }
-#' \if{latex}{
-#' \eqn{w_i} are the sample weights,
-#' \eqn{I_{i}}{I_i} is 1 if observation \eqn{i} belongs to the positive class, and 0 otherwise.
+#'    weighted.mean(((t == positive) - p)^2, w),
 #' }
+#' where \eqn{w_i} are the sample weights,
+#'  and \eqn{I_{i}} is 1 if observation \eqn{x_i} belongs to the positive class, and 0 otherwise.
 #'
 #' Note that this (more common) definition of the Brier score is equivalent to the
 #' original definition of the multi-class Brier score (see [mbrier()]) divided by 2.

diff --git a/R/binary_fbeta.R b/R/binary_fbeta.R
@@ -22,7 +22,7 @@
 #' @inheritParams binary_params
 #' @param beta (`numeric(1)`)\cr
 #'   Parameter to give either precision or recall more weight.
-#'   Default is 1, resulting in balanced weights.
+#'   Default is `1`, resulting in balanced weights.
 #' @template binary_example
 #' @export
 fbeta = function(truth, response, positive, beta = 1, na_value = NaN, ...) {

diff --git a/R/binary_fn.R b/R/binary_fn.R
@@ -3,7 +3,7 @@
 #' @details
 #' This measure counts the false negatives (type 2 error), i.e. the number of
 #' predictions indicating a negative class label while in fact it is positive.
-#' This is sometimes also called a "false alarm".
+#' This is sometimes also called a "miss" or an "underestimation".
 #'
 #' @templateVar mid fn
 #' @template binary_template

diff --git a/R/binary_fp.R b/R/binary_fp.R
@@ -3,6 +3,7 @@
 #' @details
 #' This measure counts the false positives (type 1 error), i.e. the number of
 #' predictions indicating a positive class label while in fact it is negative.
+#' This is sometimes also called a "false alarm".
 #'
 #' @templateVar mid fp
 #' @template binary_template

diff --git a/R/binary_gmean.R b/R/binary_gmean.R
@@ -2,7 +2,7 @@
 #'
 #' @details
 #' Calculates the geometric mean of [recall()] R and [specificity()] S as \deqn{
-#'    \sqrt{\mathrm{R} \mathrm{S}}.
+#'    \sqrt{\mathrm{R} \cdot \mathrm{S}}.
 #' }{
 #'    sqrt(R * S)
 #' }

diff --git a/R/binary_gpr.R b/R/binary_gpr.R
@@ -2,7 +2,7 @@
 #'
 #' @details
 #' Calculates the geometric mean of [precision()] P and [recall()] R as \deqn{
-#'    \sqrt{\mathrm{P} \mathrm{R}}.
+#'    \sqrt{\mathrm{P} \cdot \mathrm{R}}.
 #' }{
 #'    sqrt(P * R)
 #' }

diff --git a/R/binary_mcc.R b/R/binary_mcc.R
diff --git a/R/binary_prauc.R b/R/binary_prauc.R
@@ -4,7 +4,7 @@
 #' Computes the area under the Precision-Recall curve (PRC).
 #' The PRC can be interpreted as the relationship between precision and recall (sensitivity),
 #' and is considered to be a more appropriate measure for unbalanced datasets than the ROC curve.
-#' The PRC is computed by integration of the piecewise function.
+#' The AUC-PRC is computed by integration of the piecewise function.
 #'
 #' @templateVar mid prauc
 #' @template binary_template

diff --git a/R/binary_tn.R b/R/binary_tn.R
@@ -2,7 +2,7 @@
 #'
 #' @details
 #' This measure counts the true negatives, i.e. the number of
-#' predictions correctly indicating a negative class label.
+#' predictions correctly indicating a negative class label. This is sometimes also called a "correct rejection".
 #'
 #' @templateVar mid tn
 #' @template binary_template

diff --git a/R/binary_tnr.R b/R/binary_tnr.R
@@ -6,7 +6,7 @@
 #' }{
 #'    TN / (FP + TN).
 #' }
-#' Also know as "specificity".
+#' Also know as "specificity" or "selectivity".
 #'
 #' @templateVar mid tnr
 #' @template binary_template