Skip to contents

List the registered measures precise_dist() accepts

Source: R/precise_dist_list.R
precise_dist_list.Rd

Expands family keywords, aliases, and canonical names into the canonical measure names understood by precise_dist(). Only measures whose backend package is installed are returned.

Usage

precise_dist_list(dists)

Arguments

dists

A character vector of family keywords, canonical names, or aliases. An unknown name is an error. Family keywords may overlap, and overlapping expansions are not deduplicated here.

Value

A list of canonical measure names.

Family keywords

The families are all_dists, static_dists, binary_dists, nominal_dists, minkowski_dists, l1_dists, intersection_dists, inner_product_dists, squared_dists, correlation_dists, other_dists, rf_dists, kernel_dists, tsne_dists, probability_dists, time_series_dists, and transport_dists. Each also works without the _dists ending, so "static" and "static_dists" are the same request.

Name conventions

  • A trailing _phil marks a philentropy implementation rather than the proxy one. Both are registered because the two implementations can differ numerically.

  • minkowski_0.5 is proxy::dist() with power 0.5.

  • random_forest_sqrt is a randomForest proximity with mtry = sqrt(ncol(data)) and ntree = 1501.

  • laplace_1 and rbf_1 are kernlab kernels whose sigma is the 0.1 quantile from kernlab::sigest(as.matrix(x), scale = FALSE, frac = 1).

  • tsne_5 runs mmtsne::x2p() at perplexity 5 and passes the result to mmtsne::p2sp().

References

Meyer D, Buchta C (2022). proxy: Distance and Similarity Measures. R package. https://CRAN.R-project.org/package=proxy.

Drost HG (2018). Philentropy: information theory and distance quantification with R. Journal of Open Source Software, 3(26), 765. doi:10.21105/joss.00765 .

Karatzoglou A, Smola A, Hornik K, Zeileis A (2004). kernlab: an S4 package for kernel methods in R. Journal of Statistical Software, 11(9), 1–20. doi:10.18637/jss.v011.i09 .

Liaw A, Wiener M (2002). Classification and regression by randomForest. R News, 2(3), 18–22.

See also

precise_dist() to run the measures, and precise_func_fact() for parameterized measures the registry does not carry.

Author

Brian Muchmore

Examples

head(unlist(precise_dist_list("static_dists")), 10)
#>  [1] "euclidean"     "supremum"      "manhattan"     "minkowski_0.5"
#>  [5] "minkowski_1.5" "minkowski_2.5" "sorensen"      "gower"        
#>  [9] "gower_phil"    "soergel"