Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

nmfAlgorithm

Listing and Retrieving NMF Algorithms


Description

nmfAlgorithm lists access keys or retrieves NMF algorithms that are stored in registry. It allows to list

Usage

nmfAlgorithm(name = NULL, version = NULL, all = FALSE,
    ...)

Arguments

name

Access key. If not missing, it must be a single character string that is partially matched against the available algorithms in the registry. In this case, if all=FALSE (default), then the algorithm is returned as an NMFStrategy object that can be directly passed to nmf. An error is thrown if no matching algorithm is found.

If missing or NULL, then access keys of algorithms – that match the criteria version, are returned. This argument is assumed to be regular expression if all=TRUE or version is not NULL.

version

version of the algorithm(s) to retrieve. Currently only value 'R' is supported, which searched for plain R implementations.

all

a logical that indicates if all algorithm keys should be returned, including the ones from alternative algorithm versions (e.g. plain R implementations of algorithms, for which a version based on optimised C updates is used by default).

...

extra arguments passed to getNMFMethod when name is not NULL and all=FALSE. It is not used otherwise.

Value

an NMFStrategy object if name is not NULL and all=FALSE, or a named character vector that contains the access keys of the matching algorithms. The names correspond to the access key of the primary algorithm: e.g. algorithm ‘lee’ has two registered versions, one plain R (‘.R#lee’) and the other uses optimised C updates (‘lee’), which will all get named ‘lee’.

See Also

Other regalgo: canFit

Examples

# list all main algorithms
nmfAlgorithm()
# list all versions of algorithms
nmfAlgorithm(all=TRUE)
# list all plain R versions
nmfAlgorithm(version='R')

NMF

Algorithms and Framework for Nonnegative Matrix Factorization (NMF)

v0.23.0
GPL (>= 2)
Authors
Renaud Gaujoux, Cathal Seoighe
Initial release
2020-07-30

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.