feemscale: Rescale FEEM spectra to a given norm and remember the scale factor

Description

Given a norm function (typically, standard deviation), scale the intensities in FEEM objects to it and optionally remember the scale factor.

Usage

feemscale(x, ...)
  # S3 method for feem
feemscale(x, norm = sd, remember = TRUE, ...)
  # S3 method for feemcube
feemscale(x, ..., progress = FALSE)
  # S3 method for list
feemscale(x, ..., progress = FALSE)

Value

feemscale.feem: a FEEM object with intensities divided by scale factor (norm(x)) and its scale attribute multiplied by the scale factor.

feemscale.feemcube: a FEEM cube built from FEEM objects scaled as described above.

feemscale.list: a list consisting of results of

feemscale generic applied to its elements.

Arguments

x: A FEEM object, a FEEM cube object, or a list of anything compatible with feemscale generic.
norm: A function taking a numeric matrix and returning its norm. Typically, sd or sumsq.
remember: Whether to remember the scale factor. If FALSE, the scale factor in the returned object is unchanged.
...: Passed as-is to feemscale, to feemscale.feem, then to the norm function. Use this to set na.rm = TRUE for functions like sd or sumsq.
progress: Set to TRUE to enable a progress bar (implemented via txtProgressBar).

References

tools::toRd(bibentry('Article', author = c( person('Rasmus', 'Bro'), person(c('Age', 'K.'), 'Smilde') ), title = 'Centering and scaling in component analysis', journal = 'Journal of Chemometrics', volume = 17, number = 1, pages = '16-33', doi = '10.1002/cem.773', year = 2003 ))

Examples

Run this code

  feemscale(feem(matrix(1:42, 6), 1:6, 1:7))