NMFfitX1-class: Structure for Storing the Best Fit Amongst Multiple NMF Runs
In renozao/NMF: Algorithms and Framework for Nonnegative Matrix Factorization (NMF)

Description Usage Arguments Details Slots Methods (by generic) See Also Examples

This class is used to return the result from a multiple run of a single NMF algorithm performed with function nmf with the – default – option keep.all=FALSE (cf. nmf).

## S4 method for signature 'NMFfitX1'
show(object)

## S4 method for signature 'NMFfitX1'
nrun(object)

## S4 method for signature 'NMFfitX1'
consensus(object, no.attrib = FALSE)

## S4 method for signature 'NMFfitX1'
minfit(object)

## S4 method for signature 'NMFfitX1'
fit(object)

## S4 method for signature 'NMFfitX1,NMFfitX1'
nmf.equal(x, y, ...)

`object`	an object of class `NMFfitX`.
`no.attrib`	a single logical that indicates that no extra attributes should be attached to the result matrix. If `TRUE`, then the following attributes are attached: `nrun`: number of algorithm runs performed; `nbasis`: number of basis components of the fitted model; Also, in this case, the result gains an extra S3 class `NMF.consensus`.
`x, y`	objects of class `NMFfitX1`, whose fitted models are compared.
`...`	other arguments passed to subsequent calls of suitable methods, usually of the same generic.

It extends both classes NMFfitX and NMFfit, and stores a the result of the best fit in its NMFfit structure.

Beside the best fit, this class allows to hold data about the computation of the multiple runs, such as the number of runs, the CPU time used to perform all the runs, as well as the consensus matrix.

Due to the inheritance from class NMFfit, objects of class NMFfitX1 can be handled exactly as the results of single NMF run – as if only the best run had been performed.

consensus: object of class matrix used to store the consensus matrix based on all the runs.
nrun: an integer that contains the number of runs performed to compute the object.
rng1: an object that contains RNG settings used for the first run. See getRNG1.

getRNG1(object = NMFfitX1): Returns the RNG settings used to compute the first of all NMF runs, amongst which object was selected as the best fit.
show(object = NMFfitX1): Show method for objects of class NMFfitX1
nrun(object = NMFfitX1): Returns the number of NMF runs performed, amongst which object was selected as the best fit.
consensus(object = NMFfitX1): Returns the consensus matrix computed while performing all NMF runs, amongst which object was selected as the best fit.

The result is the matrix stored in slot ‘consensus’. This method returns NULL if the consensus matrix is empty.
minfit(object = NMFfitX1): Returns the fit object associated with the best fit, amongst all the runs performed when fitting object.

Since NMFfitX1 objects only hold the best fit, this method simply returns object coerced into an NMFfit object.
fit(object = NMFfitX1): Returns the model object associated with the best fit, amongst all the runs performed when fitting object.

Since NMFfitX1 objects only hold the best fit, this method simply returns the NMF model fitted by object – that is stored in slot ‘fit’.
nmf.equal(x = NMFfitX1,y = NMFfitX1): Compares the NMF models fitted by multiple runs, that only kept the best fits.

Other multipleNMF: NMFfitX-class, NMFfitXn-class

# generate a synthetic dataset with known classes
n <- 20; counts <- c(5, 2, 3);
V <- syntheticNMF(n, counts)

# get the class factor
groups <- V$pData$Group

# perform multiple runs of one algorithm, keeping only the best fit (default)
#i.e.: the implicit nmf options are .options=list(keep.all=FALSE) or .options='-k'
res <- nmf(V, 3, nrun=3) 
res

# compute summary measures
summary(res)
# get more info
summary(res, target=V, class=groups)

# show computational time
runtime.all(res)

# plot the consensus matrix, as stored (pre-computed) in the object
## Not run:  consensusmap(res, annCol=groups)