iterative_dist_remove: iteratively remove objects from a distance matrix and plot at...

In Jtrachsel/funfuns: Functions I Use (Title Case)

Description

iteratively remove objects from a distance matrix and plot at each iteration

Usage

iterative_dist_remove(
  in_dist,
  trymax = 50,
  iterations = 20,
  maxit = 1000,
  exclusion_prob = 0.95,
  parallel = 8,
  targ_pat = "SX",
  rem_type = "total",
  run_nmds = FALSE
)

Arguments

in_dist	input distance matrix, must be type dist
trymax	passed to metaMDS function
iterations	number of removal iterations to try
maxit	passed to metaMDS function
exclusion_prob	passed to quantile(), all observations with cumulative distances to targets (or total) above this quantile will be removed each iteration
parallel	passed to metaMDS function
targ_pat	passed to grep, defines the targets you are interested in (ones you dont want to remove). Only used for rem_type = 'cum_targ' and 'min'
rem_type	one of 'total', 'cum_targ', and 'min'. Specifies how to calculate distances for removal. total will pass cumulative distances to quantile, meaning the objects with the greatest cumulative distance from all other objects will be removed first. 'cum_targ' considers cumulative distances to objects selected by your targ_pat parameter, objects with greatest cumulative distances to your targets will be removed first. min considers the minimum distance to your targets, that is, the distance from each object to the closest target. probably a good choice if your targets are similar to each other and you want to keep all objects that are very similar to any of your targets.
run_nmds	should nmds be run?

Value

returns a big ol list

Examples

#soon

Jtrachsel/funfuns documentation built on Aug. 8, 2021, 7:31 p.m.