utilisation de k-means au lieu de hierarchique dans initSmallEM - PB de dimensions...

[valse.git] / R / initSmallEM.R

diff --git a/R/initSmallEM.R b/R/initSmallEM.R
index 8f3c86b..8cfb7e8 100644 (file)
--- a/R/initSmallEM.R
+++ b/R/initSmallEM.R
@@ -1,27 +1,18 @@
-library(MASS) #generalized inverse of matrix Monroe-Penrose
-
-vec_bin = function(X,r){
-  Z = c()
-  indice = c()
-  j=1
-  for(i in 1:length(X)){
-    if(X[i] == r){
-      Z[i] = 1
-      indice[j] = i
-      j=j+1
-    }
-    else{
-      Z[i] = 0
-    }
-  }
-  return(list(Z,indice))
-}
-
-initSmallEM = function(k,X,Y,tau){
+#' initialization of the EM algorithm
+#'
+#' @param k number of components
+#' @param X matrix of covariates (of size n*p)
+#' @param Y matrix of responses (of size n*m)
+#' @param tau threshold to stop EM algorithm
+#'
+#' @return a list with phiInit, rhoInit, piInit, gamInit
+#' @export
+initSmallEM = function(k,X,Y,tau)
+{
   n = nrow(Y)
   m = ncol(Y)
   p = ncol(X)
-
+  
   betaInit1 = array(0, dim=c(p,m,k,20))
   sigmaInit1 = array(0, dim = c(m,m,k,20))
   phiInit1 = array(0, dim = c(p,m,k,20))
@@ -30,28 +21,34 @@ initSmallEM = function(k,X,Y,tau){
   gamInit1 = array(0, dim=c(n,k,20))
   LLFinit1 = list()
   
-  
-  for(repet in 1:20){
-    clusters = hclust(dist(y)) #default distance : euclidean
-    clusterCut = cutree(clusters,k)
-    Zinit1[,repet] = clusterCut #retourne les indices (à quel cluster indiv_i appartient) d'un clustering hierarchique (nb de cluster = k)
+  require(MASS) #Moore-Penrose generalized inverse of matrix
+  require(mclust) # K-means with selection of K
+  for(repet in 1:20)
+  {
+    clusters = Mclust(matrix(c(X,Y),nrow=n),k) #default distance : euclidean
+    Zinit1[,repet] = clusters$classification
     
-    for(r in 1:k){
+    for(r in 1:k)
+    {
       Z = Zinit1[,repet]
       Z_bin = vec_bin(Z,r)
-      Z_vec = Z_bin[[1]] #vecteur 0 et 1 aux endroits où Z==r
-      Z_indice = Z_bin[[2]] #renvoit les indices où Z==r
+      Z_vec = Z_bin$Z #vecteur 0 et 1 aux endroits o? Z==r
+      Z_indice = Z_bin$indice #renvoit les indices o? Z==r
       
-      betaInit1[,,r,repet] = ginv(t(x[Z_indice,])%*%x[Z_indice,])%*%t(x[Z_indice,])%*%y[Z_indice,]
+      betaInit1[,,r,repet] =
+        ginv(t(x[Z_indice,])%*%x[Z_indice,])%*%t(x[Z_indice,])%*%y[Z_indice,]
       sigmaInit1[,,r,repet] = diag(m)
       phiInit1[,,r,repet] = betaInit1[,,r,repet]/sigmaInit1[,,r,repet]
       rhoInit1[,,r,repet] = solve(sigmaInit1[,,r,repet])
       piInit1[repet,r] = sum(Z_vec)/n
     }
     
-    for(i in 1:n){
-      for(r in 1:k){
-        dotProduct = (y[i,]%*%rhoInit1[,,r,repet]-x[i,]%*%phiInit1[,,r,repet]) %*% (y[i,]%*%rhoInit1[,,r,repet]-x[i,]%*%phiInit1[,,r,repet])
+    for(i in 1:n)
+    {
+      for(r in 1:k)
+      {
+        dotProduct = (y[i,]%*%rhoInit1[,,r,repet]-x[i,]%*%phiInit1[,,r,repet]) %*%
+          (y[i,]%*%rhoInit1[,,r,repet]-x[i,]%*%phiInit1[,,r,repet])
         Gam[i,r] = piInit1[repet,r]*det(rhoInit1[,,r,repet])*exp(-0.5*dotProduct)
       }
       sumGamI = sum(gam[i,])
@@ -61,20 +58,17 @@ initSmallEM = function(k,X,Y,tau){
     miniInit = 10
     maxiInit = 11
     
-    new_EMG = EMGLLF(phiInit1[,,,repet],rhoInit1[,,,repet],piInit1[repet,],gamInit1[,,repet],miniInit,maxiInit,1,0,x,y,tau)
-    ##.C("EMGLLF", phiInit = phiInit, rhoInit = rhoInit, ...)
-    LLFEessai = new_EMG[[4]]
-    LLFinit1[[repet]] = LLFEessai[[length(LLFEessai)]]
+    new_EMG = .Call("EMGLLF",phiInit1[,,,repet],rhoInit1[,,,repet],piInit1[repet,],
+                    gamInit1[,,repet],miniInit,maxiInit,1,0,x,y,tau)
+    LLFEessai = new_EMG$LLF
+    LLFinit1[repet] = LLFEessai[length(LLFEessai)]
   }
   
   b = which.max(LLFinit1)
-  
   phiInit = phiInit1[,,,b]
   rhoInit = rhoInit1[,,,b]
   piInit = piInit1[b,]
   gamInit = gamInit1[,,b]
   
-  return(list(phiInit, rhoInit, piInit, gamInit))
+  return (list(phiInit=phiInit, rhoInit=rhoInit, piInit=piInit, gamInit=gamInit))
 }
-
-