X-Git-Url: https://git.auder.net/?p=valse.git;a=blobdiff_plain;f=pkg%2FR%2Fmain.R;h=e741d65f4ed3b43037dc8978a96f20e76cda0523;hp=64e058629859e8b1442e5a1110a2eb8670b554bf;hb=ea5860f1b4fc91f06e371a0b26915198474a849d;hpb=1b698c1619dbcf5b3a0608dc894d249945d2bce3

diff --git a/pkg/R/main.R b/pkg/R/main.R
index 64e0586..e741d65 100644
--- a/pkg/R/main.R
+++ b/pkg/R/main.R
@@ -31,9 +31,9 @@ valse <- function(X, Y, procedure = "LassoMLE", selecMod = "DDSE", gamma = 1, mi
   ncores_inner = 1, thresh = 1e-08, size_coll_mod = 10, fast = TRUE, verbose = FALSE, 
   plot = TRUE)
 {
-  p <- dim(X)[2]
-  m <- dim(Y)[2]
-  n <- dim(X)[1]
+  n <- nrow(X)
+  p <- ncol(X)
+  m <- ncol(Y)
 
   if (verbose) 
     print("main loop: over all k and all lambda")
@@ -57,7 +57,7 @@ valse <- function(X, Y, procedure = "LassoMLE", selecMod = "DDSE", gamma = 1, mi
     # 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.
-    P <- initSmallEM(k, X, Y)
+    P <- initSmallEM(k, X, Y, fast)
     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) 
@@ -138,7 +138,7 @@ valse <- function(X, Y, procedure = "LassoMLE", selecMod = "DDSE", gamma = 1, mi
     for (r in 1:length(modelSel$pi))
     {
       sqNorm2 <- sum((Y[i, ] %*% modelSel$rho[, , r] - X[i, ] %*% modelSel$phi[, , r])^2)
-      Gam[i, r] <- modelSel$pi[r] * exp(-0.5 * sqNorm2) * det(modelSel$rho[, , r])
+      Gam[i, r] <- modelSel$pi[r] * exp(-0.5 * sqNorm2) * gdet(modelSel$rho[, , r])
     }
   }
   Gam <- Gam/rowSums(Gam)