Reference

Main user functions
nestorFit()	Core function of nestorFit
List.nestorFit()	Run function nestorFit on a list of initial cliques using parallel computation (mclapply)
generate_missing_data()	Simulate data and parameters including missing actors.
norm_PLN()	Runs PLN function from the PLNmodels package and normalized the outputs
initVEM()	Initialize all parameters for the variational inference
Computing functions
VEstep()	Computes the variational expectation step of the algorithm
Mstep()	Computes the maximization step of the algorithm
computeOmega()	Updates the precision terms
computeWg()	Updates the variational edges weights inside the VEM.
exactMeila()	Calculates the Meila matrix using exact computation
Kirshner()	Computes edges probability from weights W (Kirshner (07) formulas)
logSumTree()	Exact computation of matrix tree log determinant
LowerBound()	Computes the lower bound.
initOmega()	Initialize Sigma and Omega using an initial clique of neighbors of the missing actor
inverse.gmp()	Computes exact inverses using gmp package
det.fractional()	Get numerator and denominators
alphaMax()	Heuristic for an upper value of tempering parameter alpha
Clique initialization
FitSparsePCA()	Fit sparse PCA on a grid of alpha
boot_FitSparsePCA()	Finds initial cliques using a sparse PCA on bootstraps sub-samples
complement_spca()	Select the k first components and their complement as initial cliques
init_blockmodels()	Find initial cliques using blockmodels on the initial marginalized network
init_mclust()	Find initial cliques using mclust on the estimated correlation matrix
Visualizations and tools
plotPerf()	Comparaitve plot of estimated edge probabilities and original graph.
plotConv()	Plot function for nestorFit convergence
ggimage()	ggplot heatmap
ppvtpr()	Computes precision and recall statistics separating observed from hidden variables, and (FPR,FNR) for hidden variables .
auc()	wraper of auc from ROCR