A panel that does not quite agree | Rank confidence sets | Probability of membership in the top k | Selection with a guarantee | Resampling the data instead of the panel | How much is a 95% set worth? | A note on cost | Related work
Package: rankimp 1.0.0
rankimp: Consensus Ranking of Variable Importance with Uncertainty
Variable importance rankings depend on the method, the random seed and the resample used to compute them. This package treats every source of importance as a judge expressing a ranking over the predictors, and synthesises those rankings into a Kemeny median ranking with ties. Uncertainty about the consensus is quantified through bootstrap rank confidence sets, top-k probabilities and clustering of disagreeing judges.
Authors:
rankimp_1.0.0.tar.gz
rankimp_1.0.0.zip(r-4.7-any)rankimp_1.0.0.zip(r-4.6-any)rankimp_1.0.0.zip(r-4.5-any)
rankimp_1.0.0.tgz(r-4.6-any)rankimp_1.0.0.tgz(r-4.5-any)
rankimp_1.0.0.tar.gz(r-4.7-any)rankimp_1.0.0.tar.gz(r-4.6-any)
rankimp_1.0.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
rankimp/json (API)
| # Install 'rankimp' in R: |
| install.packages('rankimp', repos = c('https://agostinognasso.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/agostinognasso/rankimp/issues
Datasets:
- applications - Synthetic loan applications, with the right answer attached
Last updated from:5f77f2a4a9. Checks:9 OK. Indexed: yes.
| Target | Result | Time | Files | Syslog |
|---|---|---|---|---|
| linux-devel | OK | 228 | ||
| source / vignettes | OK | 249 | ||
| linux-release | OK | 145 | ||
| macos-release | OK | 213 | ||
| macos-oldrel | OK | 256 | ||
| windows-devel | OK | 144 | ||
| windows-release | OK | 152 | ||
| windows-oldrel | OK | 142 | ||
| wasm-release | OK | 124 |
Exports:autoplotconsensus_rankimportance_judgesimportance_locoimportance_mdiimportance_permutationimportance_shapimportance_to_rankitem_consensusjudge_clustersjudge_weightsprob_topkrank_confsetsrank_select
Dependencies:cliConsRankcpp11data.tabledplyrfarvergenericsggplot2gluegtablegtoolsisobandjsonlitelabelinglifecyclemagrittrpillarpkgconfigproxypurrrR6RColorBrewerRcpprlangrlistS7scalesstringistringrtibbletidyrtidyselectutf8vctrsviridisLitewithrXMLyaml
How stable is a variable importance ranking?
Last update: 2026-09-08
Started: 2026-07-09
Robustly important variables in credit scoring
The portfolio | The panel | What the panel agrees on | What survives resampling | Where the disagreement lives | What a consensus cannot do for you
Last update: 2026-09-08
Started: 2026-07-09
Consensus ranking of variable importance
The consensus | Starting from importance scores | Building the panel from fitted models | Ties are information | Not every judge deserves an equal vote | What comes next
Last update: 2026-09-08
Started: 2026-07-09
Every function, and when you want it
The pipeline | Which function do I want? | Getting importance out of a model | Turning scores into rankings | Building a panel | Weighting the judges | The consensus | Asking who disagrees | Uncertainty | Turning uncertainty into a decision | Plots | Under the hood | Where to go next
Last update: 2026-09-08
Started: 2026-09-08
Set stability is not rank stability
Two panels | What set stability sees | What rank consensus sees | Where the two agree | The honest summary
Last update: 2026-09-08
Started: 2026-07-10
The theory the package rests on
The problem | Why rankings and not scores | The Kemeny median | Ties are part of the answer | Measuring agreement | Uncertainty about a rank | What the intervals look like, and why | Testing whether a panel is one population | What none of this protects you from | Sources
Last update: 2026-09-08
Started: 2026-09-08
When XAI methods disagree
A dataset built to cause it | Building the panel | The two methods do disagree | But the rankings agree | Is there one panel, or two? | How often does it divide? | What a rank-based diagnostic can and cannot see
Last update: 2026-09-08
Started: 2026-07-09
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| Synthetic loan applications, with the right answer attached | applications |
| Visualise a consensus ranking against the panel it came from | autoplot.consensus_rank |
| Visualise the disagreement between judges | autoplot.judge_clusters |
| Visualise the rank confidence sets | autoplot.rank_confsets |
| Kemeny consensus ranking of variable importance | consensus_rank print.consensus_rank |
| Importance backends | importance_backends importance_loco importance_mdi importance_permutation importance_shap |
| Assemble the panel of judges | importance_judges print.judges |
| Turn importance scores into rankings | importance_to_rank |
| Agreement of each judge with the consensus | item_consensus |
| Do the judges agree? | judge_clusters print.judge_clusters |
| Weights for the panel of judges | judge_weights |
| Probability that a variable lands in the top k | prob_topk |
| Bootstrap confidence sets for the consensus ranking | print.rank_confsets rank_confsets |
| Select variables with a rank guarantee | rank_select |