pkg/R/constructionModelesLassoMLE.R

   1 #' constructionModelesLassoMLE
   2 #'
   3 #' TODO: description
   4 #'
   5 #' @param ...
   6 #'
   7 #' @return ...
   8 #'
   9 #' export
  10 constructionModelesLassoMLE = function(phiInit, rhoInit, piInit, gamInit, mini, maxi,
  11     gamma, X, Y, thresh, tau, S, ncores=3, artefact = 1e3, fast=TRUE, verbose=FALSE)
  12 {
  13     if (ncores > 1)
  14     {
  15         cl = parallel::makeCluster(ncores, outfile='')
  16         parallel::clusterExport( cl, envir=environment(),
  17             varlist=c("phiInit","rhoInit","gamInit","mini","maxi","gamma","X","Y","thresh",
  18             "tau","S","ncores","verbose") )
  19     }
  20
  21     # Individual model computation
  22     computeAtLambda <- function(lambda)
  23     {
  24         if (ncores > 1)
  25             require("valse") #nodes start with an empty environment
  26
  27         if (verbose)
  28             print(paste("Computations for lambda=",lambda))
  29
  30         n = dim(X)[1]
  31         p = dim(phiInit)[1]
  32         m = dim(phiInit)[2]
  33         k = dim(phiInit)[3]
  34
  35         sel.lambda = S[[lambda]]$selected
  36 #       col.sel = which(colSums(sel.lambda)!=0) #if boolean matrix
  37         col.sel <- which( sapply(sel.lambda,length) > 0 ) #if list of selected vars
  38
  39         if (length(col.sel) == 0)
  40             return (NULL)
  41
  42         # lambda == 0 because we compute the EMV: no penalization here
  43         res = EMGLLF(phiInit[col.sel,,],rhoInit,piInit,gamInit,mini,maxi,gamma,0,
  44             X[,col.sel], Y, tau, fast)
  45
  46         # Eval dimension from the result + selected
  47         phiLambda2 = res$phi
  48         rhoLambda = res$rho
  49         piLambda = res$pi
  50         phiLambda = array(0, dim = c(p,m,k))
  51         for (j in seq_along(col.sel))
  52             phiLambda[col.sel[j],,] = phiLambda2[j,,]
  53         dimension = length(unlist(sel.lambda))
  54
  55         # Computation of the loglikelihood
  56         densite = vector("double",n)
  57         for (r in 1:k)
  58         {
  59             delta = (Y%*%rhoLambda[,,r] - (X[, col.sel]%*%phiLambda[col.sel,,r]))/artefact
  60             densite = densite + piLambda[r] *
  61                 det(rhoLambda[,,r])/(sqrt(2*base::pi))^m * exp(-tcrossprod(delta)/2.0)
  62         }
  63         llhLambda = c( sum(artefact^2 * log(densite)), (dimension+m+1)*k-1 )
  64         list("phi"= phiLambda, "rho"= rhoLambda, "pi"= piLambda, "llh" = llhLambda)
  65     }
  66
  67     # For each lambda, computation of the parameters
  68     out =
  69         if (ncores > 1)
  70             parLapply(cl, 1:length(S), computeAtLambda)
  71         else
  72             lapply(1:length(S), computeAtLambda)
  73
  74     if (ncores > 1)
  75         parallel::stopCluster(cl)
  76
  77     out
  78 }