Last updated on 2024-12-18 19:49:44 CET.
| Package | ERROR | NOTE | OK |
|---|---|---|---|
| mlr3fairness | 1 | 10 | 2 |
Current CRAN status: ERROR: 1, NOTE: 10, OK: 2
Version: 0.3.2
Check: Rd files
Result: NOTE
checkRd: (-1) groupdiff_tau.Rd:23: Lost braces
23 | \code{groupdiff_tau()} computes \eqn{min(x/y, y/x)}, i.e. the smallest symmetric ratio between \eqn{x} and eqn{y}
| ^
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-devel-windows-x86_64, r-patched-linux-x86_64, r-release-linux-x86_64, r-release-macos-arm64, r-release-macos-x86_64, r-release-windows-x86_64
Version: 0.3.2
Check: Rd cross-references
Result: NOTE
Found the following Rd file(s) with Rd \link{} targets missing package
anchors:
MeasureFairness.Rd: mlr_measures_classif.ce,
mlr_measures_classif.fpr, mlr_measures, Task
MeasureFairnessComposite.Rd: Task
MeasureFairnessConstraint.Rd: Task
MeasureSubgroup.Rd: mlr_measures_classif.fpr
compas.Rd: TaskClassif
compute_metrics.Rd: Task
fairness_accuracy_tradeoff.Rd: PredictionClassif, BenchmarkResult,
ResampleResult, Task, Measure, TaskClassif
fairness_compare_metrics.Rd: PredictionClassif, BenchmarkResult,
ResampleResult, Measure, TaskClassif, Task
fairness_prediction_density.Rd: PredictionClassif, ResampleResult,
BenchmarkResult, Task, TaskClassif
fairness_tensor.Rd: data.table, PredictionClassif, ResampleResult,
TaskClassif, Task
groupdiff_tau.Rd: Task
groupwise_metrics.Rd: Task
mlr_learners_classif.fairfgrrm.Rd: Learner, mlr_learners, lrn
mlr_learners_classif.fairzlrm.Rd: Learner, mlr_learners, lrn
mlr_learners_fairness.Rd: Task
mlr_learners_regr.fairfrrm.Rd: Learner, mlr_learners, lrn
mlr_learners_regr.fairnclm.Rd: Learner, mlr_learners, lrn
mlr_learners_regr.fairzlm.Rd: Learner, mlr_learners, lrn
mlr_measures_fairness.Rd: Task
mlr_pipeops_equalized_odds.Rd: R6Class, PipeOpTaskPreproc, PipeOp
mlr_pipeops_explicit_pta.Rd: R6Class, PipeOpTaskPreproc, PipeOp
mlr_pipeops_reweighing.Rd: R6Class, PipeOpTaskPreproc, PipeOp
report_fairness.Rd: Task
task_summary.Rd: Task
Please provide package anchors for all Rd \link{} targets not in the
package itself and the base packages.
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-windows-x86_64
Version: 0.3.2
Check: re-building of vignette outputs
Result: ERROR
Error(s) in re-building vignettes:
--- re-building ‘debiasing-vignette.Rmd’ using rmarkdown
Pandoc is required to build R Markdown vignettes but not available. Please make sure it is installed.
2023-07-10 23:46:26.428 R[80044:2290430804] XType: Using static font registry.
Warning in file.info(x, extra_cols = FALSE) :
expanded path length 4561 would be too long for
<h1 id="introduction-fairness-pipeline-operators">Introduction: Fairness Pipeline Operators</h1>
<p>Given we detected some form of bias during bias auditing, we are often interested in obtaining fair(er) models.
There are several ways to achieve this, such as collecting additional data or finding and fixing errors in the data.
Assuming there are no biases in the data and labels, one other option is to debias models using either <strong>preprocessing</strong>, <strong>postprocessing</strong> and <strong>inprocessing</strong> methods.
<code>mlr3fairness</code> provides some operators as <code>PipeOp</code>s for <code>mlr3pipelines</code>.
If you are not familiar with <code>mlr3pipelines</code>, the <a href="https://mlr3book.mlr-org.com/pipelines.html">mlr3 book</a> contains an introduction.</p>
<p>We again showcase debiasing using the <code>adult_train</code> task:</p>
<pre><code class="language-r">library(mlr3)
library(mlr3fairness)
librar [... truncated]
--- finished re-building ‘debiasing-vignette.Rmd’
--- re-building ‘measures-vignette.Rmd’ using rmarkdown
Pandoc is required to build R Markdown vignettes but not available. Please make sure it is installed.
Warning in file.info(x, extra_cols = FALSE) :
expanded path length 17990 would be too long for
<h1 id="fairness-measures">Fairness Measures</h1>
<p>Fairness measures (or metrics) allow us to assess and audit for possible biases in a trained model.
There are several types of metrics that are widely used in order to assess a model’s fairness.
They can be coarsely classified into three groups:</p>
<ul>
<li>
<p><strong>Statistical Group Fairness Metrics</strong>: Given a set of predictions from our model, we assess for differences in one or multiple metrics across groups given by a <em>protected attribute</em> [@fairmlbook; @hardt2016equality].</p>
</li>
<li>
<p><strong>Individual Fairness</strong>: Basically requires that similar people are treated similar independent of the protected attribute [@dwork2012].
We will briefly introduce individual fairness in a dedicated section below.</p>
</li>
<li>
<p><strong>Causal Fairness Notions</strong>: An important realization in the context of Fairness is, that whether a process is fair is o [... truncated]
--- finished re-building ‘measures-vignette.Rmd’
--- re-building ‘reports-vignette.Rmd’ using rmarkdown
Pandoc is required to build R Markdown vignettes but not available. Please make sure it is installed.
Quitting from lines 52-54 [build_modelcard_example_for_vignette] (reports-vignette.Rmd)
Error: processing vignette 'reports-vignette.Rmd' failed with diagnostics:
pandoc version 1.12.3 or higher is required and was not found (see the help page ?rmarkdown::pandoc_available).
--- failed re-building ‘reports-vignette.Rmd’
--- re-building ‘visualization-vignette.Rmd’ using rmarkdown
Pandoc is required to build R Markdown vignettes but not available. Please make sure it is installed.
2023-07-10 23:48:42.299 R[18209:2290505996] XType: Using static font registry.
Warning in file.info(x, extra_cols = FALSE) :
expanded path length 8617 would be too long for
<pre><code class="language-r">library(mlr3)
library(mlr3fairness)
library(mlr3learners)
</code></pre>
<h1 id="why-we-need-fairness-visualizations">Why we need fairness visualizations:</h1>
<p>Through fairness visualizations allow for first investigations into possible fairness problems in a dataset.
In this vignette we will showcase some of the pre-built fairness visualization functions.
All the methods showcased below can be used together with objects of type <code>BenchmarkResult</code>, <code>ResampleResult</code> and <code>Prediction</code>.</p>
<h1 id="the-scenario">The scenario</h1>
<p>For this example, we use the <code>adult_train</code> dataset.
Keep in mind all the datasets from <code>mlr3fairness</code> package already set protected attribute via the <code>col_role</code> “pta”, here the “sex” column.</p>
<pre><code class="language-r">t = tsk("adult_train")
t$col_roles$pta
#> [1] "sex"
</code></pre [... truncated]
--- finished re-building ‘visualization-vignette.Rmd’
SUMMARY: processing the following file failed:
‘reports-vignette.Rmd’
Error: Vignette re-building failed.
Execution halted
Flavor: r-oldrel-macos-x86_64