X-Git-Url: https://git.auder.net/?a=blobdiff_plain;f=epclust%2FR%2Fmain.R;h=892c64c631ffafc67058e5d491ba6e86923ca4ba;hb=95b5c2e621af8949c5a5eed287d451817c16c24e;hp=bf31c87d6f18ed2e8b6f226a17859e0d098253dd;hpb=af3ea94782db30c2fc1bd796d788dab5e1fb79ed;p=epclust.git

diff --git a/epclust/R/main.R b/epclust/R/main.R
index bf31c87..892c64c 100644
--- a/epclust/R/main.R
+++ b/epclust/R/main.R
@@ -144,7 +144,7 @@ claws = function(getSeries, K1, K2,
 		getSeries = function(inds) getDataInFile(inds, series_file, nbytes, endian)
 	}
 
-	# Serialize all computed wavelets contributions onto a file
+	# Serialize all computed wavelets contributions into a file
 	contribs_file = paste(bin_dir,"contribs",sep="") ; unlink(contribs_file)
 	index = 1
 	nb_curves = 0
@@ -161,22 +161,6 @@ claws = function(getSeries, K1, K2,
 	if (nb_series_per_task < min_series_per_chunk)
 		stop("Too many tasks: less series in one task than min_series_per_chunk!")
 
-	# Cluster contributions in parallel (by nb_series_per_chunk)
-	indices_all = if (random) sample(nb_curves) else seq_len(nb_curves)
-	indices_tasks = lapply(seq_len(ntasks), function(i) {
-		upper_bound = ifelse( i<ntasks, min(nb_series_per_task*i,nb_curves), nb_curves )
-		indices_all[((i-1)*nb_series_per_task+1):upper_bound]
-	})
-	if (verbose)
-		cat(paste("...Run ",ntasks," x stage 1 in parallel\n",sep=""))
-	if (parll)
-	{
-		cl = parallel::makeCluster(ncores_tasks)
-		parallel::clusterExport(cl, varlist=c("getSeries","getContribs","K1","K2","verbose","parll",
-			"nb_series_per_chunk","ncores_clust","synchrones_file","sep","nbytes","endian"),
-			envir = environment())
-	}
-
 	runTwoStepClustering = function(inds)
 	{
 		if (parll)
@@ -185,6 +169,15 @@ claws = function(getSeries, K1, K2,
 			inds, getContribs, K1, nb_series_per_chunk, ncores_clust, verbose, parll)
 		if (WER=="mix")
 		{
+
+
+
+
+#TODO: getSeries(indices_medoids) BAD ; il faudrait une big.matrix de medoids en entree
+			#OK en RAM il y en aura 1000 (donc 1000*K1*17519... OK)
+			#...mais du coup chaque process ne re-dupliquera pas medoids
+
+
 			medoids2 = computeClusters2(getSeries(indices_medoids),
 				K2, getSeries, nb_curves, nb_series_per_chunk, ncores_clust, verbose, parll)
 			binarize(medoids2, synchrones_file, nb_series_per_chunk, sep, nbytes, endian)
@@ -193,6 +186,26 @@ claws = function(getSeries, K1, K2,
 		indices_medoids
 	}
 
+	# Cluster contributions in parallel (by nb_series_per_chunk)
+	indices_all = if (random) sample(nb_curves) else seq_len(nb_curves)
+	indices_tasks = lapply(seq_len(ntasks), function(i) {
+		upper_bound = ifelse( i<ntasks, min(nb_series_per_task*i,nb_curves), nb_curves )
+		indices_all[((i-1)*nb_series_per_task+1):upper_bound]
+	})
+	if (verbose)
+		cat(paste("...Run ",ntasks," x stage 1 in parallel\n",sep=""))
+	if (WER=="mix")
+		{synchrones_file = paste(bin_dir,"synchrones",sep="") ; unlink(synchrones_file)}
+	if (parll)
+	{
+		cl = parallel::makeCluster(ncores_tasks)
+		varlist = c("getSeries","getContribs","K1","K2","verbose","parll",
+			"nb_series_per_chunk","ncores_clust","sep","nbytes","endian")
+		if (WER=="mix")
+			varlist = c(varlist, "synchrones_file")
+		parallel::clusterExport(cl, varlist=varlist, envir = environment())
+	}
+
 	# 1000*K1 indices [if WER=="end"], or empty vector [if WER=="mix"] --> series on file
 	if (parll)
 		indices = unlist( parallel::parLapply(cl, indices_tasks, runTwoStepClustering) )
@@ -202,7 +215,6 @@ claws = function(getSeries, K1, K2,
 		parallel::stopCluster(cl)
 
 	getRefSeries = getSeries
-	synchrones_file = paste(bin_dir,"synchrones",sep="") ; unlink(synchrones_file)
 	if (WER=="mix")
 	{
 		indices = seq_len(ntasks*K2)
@@ -218,6 +230,10 @@ claws = function(getSeries, K1, K2,
 			contribs_file, nb_series_per_chunk, nbytes, endian)
 	}
 
+
+
+#TODO: if ntasks==1, c'est deja terminé
+
 	# Run step2 on resulting indices or series (from file)
 	if (verbose)
 		cat("...Run final // stage 1 + stage 2\n")