nestorFit: Core function of nestorFit

In Rmomal/nestor: Network inference from Species counTs with missing actORs

Description

Core function of nestorFit

Usage

nestorFit(
  MO,
  SO,
  initList,
  maxIter = 20,
  eps = 0.01,
  alpha = 0.1,
  verbatim = 1,
  print.hist = FALSE,
  trackJ = FALSE
)

Arguments

MO	Estimated means from norm_PLN.
SO	Estimated marginal vairances from norm_PLN.
initList	Result list from initVEM.
maxIter	Maximal number of iterations.
eps	Convergence precision parameter.
alpha	Tempering parameter, default to 0.1.
verbatim	Integer controlling verbosity in three levels, starting at 0.
print.hist	Prints edges weights histograms at each step if TRUE.
trackJ	Computes the lower bound at each parameter update if TRUE. Otherwise, the lower bound is only computed at each new VE step.

Value

• M: estimated means.

• S: estimated marginal variances.

• Pg: edges probabilities.

• Wg: edges variational weights.

• W: edges weights.

• Omega: matrix filled with precision terms common to all spanning trees.

• lowbound: table containing the lowerbound trajectory.

• features: table containing the parametes trajectory.

• finalIter: number of iterations until convergence was reached.

• time: running time of the VEM.

• max.prec: boolean for reach of maximal precision reached during the VEM fit.

Examples

data=generate_missing_data(n=100,p=10,r=1,type="scale-free", plot=FALSE)
PLNfit<-norm_PLN(data$Y)
MO<-PLNfit$MO
SO<-PLNfit$SO
sigma_O=PLNfit$sigma_O
#-- initialize with true clique for example
initClique=data$TC
#-- initialize the VEM
initList=initVEM(cliqueList=initClique,sigma_O, MO,r=1 )
#-- run core function nestorFit
fit=nestorFit( MO,SO, initList=initList, maxIter=5,verbatim=1)
str(fit)

Rmomal/nestor documentation built on Dec. 16, 2020, 7:54 p.m.