#' @param kmax integer, maximum number of clusters, by default = 10
#' @param rang.min integer, minimum rank in the low rank procedure, by default = 1
#' @param rang.max integer, maximum rank in the
+#' @param ncores_outer Number of cores for the outer loop on k
+#' @param ncores_inner Number of cores for the inner loop on lambda
+#' @param size_coll_mod (Maximum) size of a collection of models
+#' @param fast TRUE to use compiled C code, FALSE for R code only
+#' @param verbose TRUE to show some execution traces
#'
#' @return a list with estimators of parameters
#'
#' @examples
#' #TODO: a few examples
#' @export
-valse = function(X,Y,procedure = 'LassoMLE',selecMod = 'DDSE',gamma = 1,mini = 10,
- maxi = 50,eps = 1e-4,kmin = 2,kmax = 2,
- rang.min = 1,rang.max = 10)
+valse = function(X, Y, procedure='LassoMLE', selecMod='DDSE', gamma=1, mini=10, maxi=50,
+ eps=1e-4, kmin=2, kmax=4, rang.min=1, rang.max=10, ncores_outer=1, ncores_inner=1,
+ size_coll_mod=50, fast=TRUE, verbose=FALSE)
{
- ####################
- # compute all models
- ####################
-
p = dim(X)[2]
m = dim(Y)[2]
n = dim(X)[1]
-
- model = list()
- tableauRecap = array(0, dim=c(1000,4))
- cpt = 0
- print("main loop: over all k and all lambda")
-
- for (k in kmin:kmax)
+
+ if (verbose)
+ print("main loop: over all k and all lambda")
+
+ if (ncores_outer > 1)
+ {
+ cl = parallel::makeCluster(ncores_outer, outfile='')
+ parallel::clusterExport( cl=cl, envir=environment(), varlist=c("X","Y","procedure",
+ "selecMod","gamma","mini","maxi","eps","kmin","kmax","rang.min","rang.max",
+ "ncores_outer","ncores_inner","verbose","p","m") )
+ }
+
+ # Compute models with k components
+ computeModels <- function(k)
{
- print(k)
- print("Parameters initialization")
- #smallEM initializes parameters by k-means and regression model in each component,
+ if (ncores_outer > 1)
+ require("valse") #nodes start with an empty environment
+
+ if (verbose)
+ print(paste("Parameters initialization for k =",k))
+ #smallEM initializes parameters by k-means and regression model in each component,
#doing this 20 times, and keeping the values maximizing the likelihood after 10
#iterations of the EM algorithm.
- init = initSmallEM(k, X, Y)
- phiInit <- init$phiInit
- rhoInit <- init$rhoInit
- piInit <- init$piInit
- gamInit <- init$gamInit
- grid_lambda <- computeGridLambda(phiInit, rhoInit, piInit, gamInit, X, Y, gamma, mini, maxi, eps)
-
- if (length(grid_lambda)>100)
- grid_lambda = grid_lambda[seq(1, length(grid_lambda), length.out = 100)]
- print("Compute relevant parameters")
+ P = initSmallEM(k, X, Y)
+ grid_lambda <- computeGridLambda(P$phiInit, P$rhoInit, P$piInit, P$gamInit, X, Y,
+ gamma, mini, maxi, eps, fast)
+ if (length(grid_lambda)>size_coll_mod)
+ grid_lambda = grid_lambda[seq(1, length(grid_lambda), length.out = size_coll_mod)]
+
+ if (verbose)
+ print("Compute relevant parameters")
#select variables according to each regularization parameter
- #from the grid: A1 corresponding to selected variables, and
- #A2 corresponding to unselected variables.
-
- params = selectiontotale(phiInit,rhoInit,piInit,gamInit,mini,maxi,gamma,grid_lambda,X,Y,1e-8,eps)
- #params2 = selectVariables(phiInit,rhoInit,piInit,gamInit,mini,maxi,gamma,grid_lambda[seq(1,length(grid_lambda), by=3)],X,Y,1e-8,eps)
- ## etrange : params et params 2 sont différents ...
- selected <- params$selected
- Rho <- params$Rho
- Pi <- params$Pi
+ #from the grid: S$selected corresponding to selected variables
+ S = selectVariables(P$phiInit, P$rhoInit, P$piInit, P$gamInit, mini, maxi, gamma,
+ grid_lambda, X, Y, 1e-8, eps, ncores_inner, fast) #TODO: 1e-8 as arg?! eps?
if (procedure == 'LassoMLE')
{
- print('run the procedure Lasso-MLE')
+ if (verbose)
+ print('run the procedure Lasso-MLE')
#compute parameter estimations, with the Maximum Likelihood
#Estimator, restricted on selected variables.
- model[[k]] = constructionModelesLassoMLE(phiInit, rhoInit,piInit,gamInit,mini,maxi,gamma,X,Y,thresh,eps,selected)
- llh = matrix(ncol = 2)
- for (l in seq_along(model[[k]]))
- llh = rbind(llh, model[[k]][[l]]$llh)
- LLH = llh[-1,1]
- D = llh[-1,2]
+ models <- constructionModelesLassoMLE(P$phiInit, P$rhoInit, P$piInit, P$gamInit,
+ mini, maxi, gamma, X, Y, thresh, eps, S, ncores_inner, artefact=1e3, fast, verbose)
}
else
{
- print('run the procedure Lasso-Rank')
+ if (verbose)
+ print('run the procedure Lasso-Rank')
#compute parameter estimations, with the Low Rank
#Estimator, restricted on selected variables.
- model = constructionModelesLassoRank(Pi, Rho, mini, maxi, X, Y, eps,
- A1, rank.min, rank.max)
-
- ################################################
- ### Regarder la SUITE
- phi = runProcedure2()$phi
- Phi2 = Phi
- if (dim(Phi2)[1] == 0)
- Phi[, , 1:k,] <- phi
- else
- {
- Phi <- array(0, dim = c(p, m, kmax, dim(Phi2)[4] + dim(phi)[4]))
- Phi[, , 1:(dim(Phi2)[3]), 1:(dim(Phi2)[4])] <<- Phi2
- Phi[, , 1:k,-(1:(dim(Phi2)[4]))] <<- phi
- }
+ models <- constructionModelesLassoRank(S$Pi, S$Rho, mini, maxi, X, Y, eps, A1,
+ rank.min, rank.max, ncores_inner, fast, verbose)
}
- tableauRecap[(cpt+1):(cpt+length(model[[k]])), ] = matrix(c(LLH, D, rep(k, length(model[[k]])), 1:length(model[[k]])), ncol = 4)
- cpt = cpt+length(model[[k]])
+ #attention certains modeles sont NULL après selectVariables
+ models = models[sapply(models, function(cell) !is.null(cell))]
+ models
}
- print('Model selection')
- tableauRecap = tableauRecap[rowSums(tableauRecap[, 2:4])!=0,]
- tableauRecap = tableauRecap[(tableauRecap[,1])!=Inf,]
- data = cbind(1:dim(tableauRecap)[1], tableauRecap[,2], tableauRecap[,2], tableauRecap[,1])
- require(capushe)
- modSel = capushe(data, n)
+
+ # List (index k) of lists (index lambda) of models
+ models_list <-
+ if (ncores_outer > 1)
+ parLapply(cl, kmin:kmax, computeModels)
+ else
+ lapply(kmin:kmax, computeModels)
+ if (ncores_outer > 1)
+ parallel::stopCluster(cl)
+
+ if (! requireNamespace("capushe", quietly=TRUE))
+ {
+ warning("'capushe' not available: returning all models")
+ return (models_list)
+ }
+
+ # Get summary "tableauRecap" from models
+ tableauRecap = do.call( rbind, lapply( seq_along(models_list), function(i) {
+ models <- models_list[[i]]
+ #Pour un groupe de modeles (même k, différents lambda):
+ LLH <- sapply( models, function(model) model$llh[1] )
+ k = length(models[[1]]$pi)
+ # TODO: chuis pas sûr du tout des lignes suivantes...
+ # J'ai l'impression qu'il manque des infos
+ ## C'est surtout que la pénalité est la mauvaise, la c'est celle du Lasso, nous on veut ici
+ ##celle de l'heuristique de pentes
+ #sumPen = sapply( models, function(model)
+ # sum( model$pi^gamma * sapply(1:k, function(r) sum(abs(model$phi[,,r]))) ) )
+ sumPen = sapply(models, function(model)
+ k*(dim(model$rho)[1]+sum(model$phi[,,1]!=0)+1)-1)
+ data.frame(model=paste(i,".",seq_along(models),sep=""),
+ pen=sumPen/n, complexity=sumPen, contrast=LLH)
+ } ) )
+print(tableauRecap)
+ modSel = capushe::capushe(tableauRecap, n)
indModSel <-
if (selecMod == 'DDSE')
as.numeric(modSel@DDSE@model)
modSel@BIC_capushe$model
else if (selecMod == 'AIC')
modSel@AIC_capushe$model
- model[[tableauRecap[indModSel,3]]][[tableauRecap[indModSel,4]]]
+
+ mod = as.character(tableauRecap[indModSel,1])
+ listMod = as.integer(unlist(strsplit(mod, "[.]")))
+ models_list[[listMod[1]]][[listMod[2]]]
+ models_list
}