co_heaviness: Co-heaviness for pairs of parent packages

View source: R/stat.R

co_heavinessR Documentation

Co-heaviness for pairs of parent packages

Description

Co-heaviness for pairs of parent packages

Usage

co_heaviness(x, rel = FALSE, a = 10, jaccard = FALSE)

Arguments

x

An object returned by pkgndep.

rel

Whether to return the absolute measure or the relative measure.

a

A constant added for calculating the relative measure.

jaccard

Whether to return Jaccard coeffcient?

Details

Denote a package as P and its two strong parent packages as A and B, i.e., parent packages in "Depends", "Imports" and "LinkingTo", the co-heaviness for A and B is calculated as follows.

Denote S_A as the set of reduced dependency packages when only moving A to "Suggests" of P, and denote S_B as the set of reduced dependency packages when only moving B to "Suggests" of P, denote S_AB as the set of reduced dependency packages when moving A and B together to "Suggests" of P, the co-heaviness of A, B on P is calculatd as length(setdiff(S_AB, union(S_A, S_B))), which is the number of reduced package only caused by co-action of A and B.

Note the co-heaviness is only calculated for parent packages in "Depends", "Imports" and "LinkingTo".

When jaccard is set to TRUE, the function returns jaccard coeffcient. setdiff(S_AB, union(S_A, S_B)) is actually the set of dependencies imported by and only by two parent packages A and B. Thus the jaccard coeffcient is calculated as length(setdiff(S_AB, union(S_A, S_B)))/length(S_AB).

Examples

## Not run: 
x = pkgndep("DESeq2")
hm = co_heaviness(x)
ComplexHeatmap::Heatmap(hm)
co_heaviness(x, jaccard = TRUE)

## End(Not run)

jokergoo/pkgndep documentation built on Nov. 24, 2022, 2:34 a.m.