Package: RAFS 0.2.5

Radosław Piliszek

RAFS: Robust Aggregative Feature Selection

A cross-validated minimal-optimal feature selection algorithm. It utilises popularity counting, hierarchical clustering with feature dissimilarity measures, and prefiltering with all-relevant feature selection method to obtain the minimal-optimal set of features.

Authors:Radosław Piliszek [aut, cre], Witold Remigiusz Rudnicki [ths, aut]

RAFS_0.2.5.tar.gz
RAFS_0.2.5.zip(r-4.7-any)RAFS_0.2.5.zip(r-4.6-any)RAFS_0.2.5.zip(r-4.5-any)
RAFS_0.2.5.tgz(r-4.6-any)RAFS_0.2.5.tgz(r-4.5-any)
RAFS_0.2.5.tar.gz(r-4.7-any)RAFS_0.2.5.tar.gz(r-4.6-any)
RAFS_0.2.5.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
RAFS/json (API)

# Install 'RAFS' in R:
install.packages('RAFS', repos = c('https://cranhaven.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/cranhaven/cranhaven.r-universe.dev/issues

On CRAN:

Conda:

archivedpackagesr-universe

1.70 score 5 stars 351 downloads 22 exports 3 dependencies

Last updated from:141a8c478c (on package/RAFS). Checks:9 OK. Indexed: no.

TargetResultTimeFilesSyslog
linux-develOK151
source / vignettesOK143
linux-releaseOK183
macos-releaseOK148
macos-oldrelOK134
windows-develOK140
windows-releaseOK143
windows-oldrelOK155
wasm-releaseOK107

Exports:builtin_dist_funscompute_fs_resultscor_distcreate_seeded_foldsdefault_dist_funsdefault_fs_fundefault_hclust_methodsget_rafs_all_reps_from_popcntsget_rafs_occurrence_matrixget_rafs_rep_tuples_matrixget_rafs_rep_tuples_popcntsget_rafs_reps_popcntsget_rafs_top_rep_tuples_from_popcntsget_rafs_top_reps_from_popcntsget_rafs_tops_popcntsget_run_idrun_rafsrun_rafs_with_fs_resultsstig_diststig_from_ig_diststig_stable_distvi_dist

Dependencies:fastclusterMDFSsplitTools

Readme and manuals

Help Manual

Help pageTopics
All built-in feature dissimilarity functionsbuiltin_dist_funs
Compute preliminary feature selection results for RAFScompute_fs_results
Feature dissimilarity based on Pearson's Correlation (cor)cor_dist
Create seeded foldscreate_seeded_folds
Default feature dissimilarity functionsdefault_dist_funs
Default (example) feature selection function for RAFSdefault_fs_fun
Default hclust methodsdefault_hclust_methods
Get all representatives from their popcntsget_rafs_all_reps_from_popcnts
Get co-occurrence matrix from RAFS resultsget_rafs_occurrence_matrix
Get representatives' tuples' co-representation matrix from RAFS resultsget_rafs_rep_tuples_matrix
Get representatives' tuples' popularity counts (popcnts) from RAFS resultsget_rafs_rep_tuples_popcnts
Get representatives' popularity counts (popcnts) from RAFS resultsget_rafs_reps_popcnts
Get top (i.e., most common) representatives's tuples from their popcntsget_rafs_top_rep_tuples_from_popcnts
Get top (i.e., most common) representatives from their popcntsget_rafs_top_reps_from_popcnts
Get top popularity counts (popcnts) from FS resultsget_rafs_tops_popcnts
Generate CV run identifiersget_run_id
Robust Aggregative Feature Selection (RAFS)run_rafs
Robust Aggregative Feature Selection (RAFS) from feature selection resultsrun_rafs_with_fs_results
Symmetric Target Information Gain (STIG) computed directlystig_dist
Symmetric Target Information Gain (STIG) computed from single Information Gains (IGs)stig_from_ig_dist
Symmetric Target Information Gain (STIG) computed directly but with pre-computed 1D conditional entropy (aka stable)stig_stable_dist
Variation of Information (VI)vi_dist