X-Git-Url: https://git.auder.net/?a=blobdiff_plain;f=epclust%2FR%2Fclustering.R;h=a431ba86e6120aef106fdce8712eb52a5ef3c4b0;hb=9f05a4a0b703deffd7bdb9cd99b0aaa2246a5c83;hp=3993e7685c97b194644d3600ceeea6b7bdac54ac;hpb=24ed5d835e2eebaaa4d5f8296f8d2e2132cc6398;p=epclust.git

diff --git a/epclust/R/clustering.R b/epclust/R/clustering.R
index 3993e76..a431ba8 100644
--- a/epclust/R/clustering.R
+++ b/epclust/R/clustering.R
@@ -1,58 +1,59 @@
 #' @name clustering
 #' @rdname clustering
-#' @aliases clusteringTask1 computeClusters1 computeClusters2
+#' @aliases clusteringTask1 clusteringTask2 computeClusters1 computeClusters2
 #'
 #' @title Two-stage clustering, withing one task (see \code{claws()})
 #'
 #' @description \code{clusteringTask1()} runs one full stage-1 task, which consists in
 #'   iterated stage 1 clustering (on nb_curves / ntasks energy contributions, computed
-#'   through discrete wavelets coefficients). \code{computeClusters1()} and
-#'   \code{computeClusters2()} correspond to the atomic clustering procedures respectively
-#'   for stage 1 and 2. The former applies the clustering algorithm (PAM) on a
-#'   contributions matrix, while the latter clusters a chunk of series inside one task
-#'   (~max nb_series_per_chunk)
+#'   through discrete wavelets coefficients).
+#'   \code{clusteringTask2()} runs a full stage-2 task, which consists in synchrones
+#'   and then WER distances computations, before applying the clustering algorithm.
+#'   \code{computeClusters1()} and \code{computeClusters2()} correspond to the atomic
+#'   clustering procedures respectively for stage 1 and 2. The former applies the
+#'   first clustering algorithm on a contributions matrix, while the latter clusters
+#'   a set of series inside one task (~nb_items_clust1)
 #'
 #' @param indices Range of series indices to cluster in parallel (initial data)
 #' @param getContribs Function to retrieve contributions from initial series indices:
 #'   \code{getContribs(indices)} outpus a contributions matrix
-#' @param contribs matrix of contributions (e.g. output of \code{curvesToContribs()})
 #' @inheritParams computeSynchrones
 #' @inheritParams claws
 #'
-#' @return For \code{clusteringTask1()} and \code{computeClusters1()}, the indices of the
-#'   computed (K1) medoids. Indices are irrelevant for stage 2 clustering, thus
-#'   \code{computeClusters2()} outputs a big.matrix of medoids
-#'   (of size limited by nb_series_per_chunk)
+#' @return For \code{clusteringTask1()}, the indices of the computed (K1) medoids.
+#'   Indices are irrelevant for stage 2 clustering, thus \code{clusteringTask2()}
+#'   outputs a big.matrix of medoids (of size LxK2, K2 = final number of clusters)
 NULL
 
 #' @rdname clustering
 #' @export
-clusteringTask1 = function(
-	indices, getContribs, K1, nb_series_per_chunk, ncores_clust=1, verbose=FALSE, parll=TRUE)
+clusteringTask1 = function(indices, getContribs, K1, algoClust1, nb_items_clust1,
+	ncores_clust=1, verbose=FALSE, parll=TRUE)
 {
-	if (verbose)
-		cat(paste("*** Clustering task on ",length(indices)," lines\n", sep=""))
-
-	wrapComputeClusters1 = function(inds) {
-		if (parll)
-			require("epclust", quietly=TRUE)
-		if (verbose)
-			cat(paste("   computeClusters1() on ",length(inds)," lines\n", sep=""))
-		inds[ computeClusters1(getContribs(inds), K1) ]
-	}
-
 	if (parll)
 	{
-		cl = parallel::makeCluster(ncores_clust)
+		cl = parallel::makeCluster(ncores_clust, outfile = "")
 		parallel::clusterExport(cl, varlist=c("getContribs","K1","verbose"), envir=environment())
 	}
 	while (length(indices) > K1)
 	{
-		indices_workers = .spreadIndices(indices, nb_series_per_chunk)
-		if (parll)
-			indices = unlist( parallel::parLapply(cl, indices_workers, wrapComputeClusters1) )
-		else
-			indices = unlist( lapply(indices_workers, wrapComputeClusters1) )
+		indices_workers = .spreadIndices(indices, nb_items_clust1)
+		if (verbose)
+			cat(paste("*** [iterated] Clustering task 1 on ",length(indices)," series\n", sep=""))
+		indices <-
+			if (parll)
+			{
+				unlist( parallel::parLapply(cl, indices_workers, function(inds) {
+					require("epclust", quietly=TRUE)
+					inds[ algoClust1(getContribs(inds), K1) ]
+				}) )
+			}
+			else
+			{
+				unlist( lapply(indices_workers, function(inds)
+					inds[ algoClust1(getContribs(inds), K1) ]
+				) )
+			}
 	}
 	if (parll)
 		parallel::stopCluster(cl)
@@ -62,19 +63,20 @@ clusteringTask1 = function(
 
 #' @rdname clustering
 #' @export
-computeClusters1 = function(contribs, K1)
-	cluster::pam(contribs, K1, diss=FALSE)$id.med
-
-#' @rdname clustering
-#' @export
-computeClusters2 = function(medoids, K2,
-	getRefSeries, nb_ref_curves, nb_series_per_chunk, ncores_clust=1,verbose=FALSE,parll=TRUE)
+clusteringTask2 = function(medoids, K2, algoClust2, getRefSeries, nb_ref_curves,
+	nb_series_per_chunk, sync_mean, nbytes,endian,ncores_clust=1,verbose=FALSE,parll=TRUE)
 {
-	synchrones = computeSynchrones(medoids,
-		getRefSeries, nb_ref_curves, nb_series_per_chunk, ncores_clust, verbose, parll)
-	distances = computeWerDists(synchrones, ncores_clust, verbose, parll)
-	#TODO: if PAM cannot take big.matrix in input, cast it before... (more than OK in RAM)
-	medoids[ cluster::pam(distances, K2, diss=TRUE)$medoids , ]
+	if (verbose)
+		cat(paste("*** Clustering task 2 on ",ncol(medoids)," synchrones\n", sep=""))
+
+	if (ncol(medoids) <= K2)
+		return (medoids)
+	synchrones = computeSynchrones(medoids, getRefSeries, nb_ref_curves,
+		nb_series_per_chunk, sync_mean, ncores_clust, verbose, parll)
+	distances = computeWerDists(synchrones, nbytes, endian, ncores_clust, verbose, parll)
+	if (verbose)
+		cat(paste("   algoClust2() on ",nrow(distances)," items\n", sep=""))
+	medoids[ ,algoClust2(distances,K2) ]
 }
 
 #' computeSynchrones
@@ -88,78 +90,96 @@ computeClusters2 = function(medoids, K2,
 #' @param nb_ref_curves How many reference series? (This number is known at this stage)
 #' @inheritParams claws
 #'
-#' @return A big.matrix of size K1 x L where L = data_length
+#' @return A big.matrix of size L x K1 where L = length of a serie
 #'
 #' @export
-computeSynchrones = function(medoids, getRefSeries,
-	nb_ref_curves, nb_series_per_chunk, ncores_clust=1,verbose=FALSE,parll=TRUE)
+computeSynchrones = function(medoids, getRefSeries, nb_ref_curves,
+	nb_series_per_chunk, sync_mean, ncores_clust=1,verbose=FALSE,parll=TRUE)
 {
-
-
-
-#TODO: si parll, getMedoids + serialization, pass only getMedoids to nodes
-# --> BOF... chaque node chargera tous les medoids (efficacité) :/ ==> faut que ça tienne en RAM
-#au pire :: C-ifier et charger medoids 1 by 1...
-
-	#MIEUX :: medoids DOIT etre une big.matrix partagée !
-
 	computeSynchronesChunk = function(indices)
 	{
-		if (verbose)
-			cat(paste("--- Compute synchrones for ",length(indices)," lines\n", sep=""))
+		if (parll)
+		{
+			require("bigmemory", quietly=TRUE)
+			requireNamespace("synchronicity", quietly=TRUE)
+			require("epclust", quietly=TRUE)
+			synchrones <- bigmemory::attach.big.matrix(synchrones_desc)
+			if (sync_mean)
+				counts <- bigmemory::attach.big.matrix(counts_desc)
+			medoids <- bigmemory::attach.big.matrix(medoids_desc)
+			m <- synchronicity::attach.mutex(m_desc)
+		}
+
 		ref_series = getRefSeries(indices)
-		#get medoids indices for this chunk of series
-		for (i in seq_len(nrow(ref_series)))
+		nb_series = ncol(ref_series)
+
+		# Get medoids indices for this chunk of series
+		mi = computeMedoidsIndices(medoids@address, ref_series)
+
+		for (i in seq_len(nb_series))
 		{
-			j = which.min( rowSums( sweep(medoids, 2, ref_series[i,], '-')^2 ) )
 			if (parll)
 				synchronicity::lock(m)
-			synchrones[j,] = synchrones[j,] + ref_series[i,]
-			counts[j,1] = counts[j,1] + 1
+			synchrones[, mi[i] ] = synchrones[, mi[i] ] + ref_series[,i]
+			if (sync_mean)
+				counts[ mi[i] ] = counts[ mi[i] ] + 1
 			if (parll)
 				synchronicity::unlock(m)
 		}
 	}
 
-	K = nrow(medoids)
+	K = ncol(medoids) ; L = nrow(medoids)
 	# Use bigmemory (shared==TRUE by default) + synchronicity to fill synchrones in //
 	# TODO: if size > RAM (not our case), use file-backed big.matrix
-	synchrones = bigmemory::big.matrix(nrow=K,ncol=ncol(medoids),type="double",init=0.)
-	counts = bigmemory::big.matrix(nrow=K,ncol=1,type="double",init=0)
+	synchrones = bigmemory::big.matrix(nrow=L, ncol=K, type="double", init=0.)
+	if (sync_mean)
+		counts = bigmemory::big.matrix(nrow=K, ncol=1, type="double", init=0)
 	# synchronicity is only for Linux & MacOS; on Windows: run sequentially
 	parll = (requireNamespace("synchronicity",quietly=TRUE)
 		&& parll && Sys.info()['sysname'] != "Windows")
-	if (parll)
-		m <- synchronicity::boost.mutex()
-
 	if (parll)
 	{
+		m <- synchronicity::boost.mutex()
+		m_desc <- synchronicity::describe(m)
+		synchrones_desc = bigmemory::describe(synchrones)
+		if (sync_mean)
+			counts_desc = bigmemory::describe(counts)
+		medoids_desc = bigmemory::describe(medoids)
 		cl = parallel::makeCluster(ncores_clust)
-		parallel::clusterExport(cl,
-			varlist=c("synchrones","counts","verbose","medoids","getRefSeries"),
-			envir=environment())
+		varlist=c("synchrones_desc","sync_mean","m_desc","medoids_desc","getRefSeries")
+		if (sync_mean)
+			varlist = c(varlist, "counts_desc")
+		parallel::clusterExport(cl, varlist, envir=environment())
 	}
 
+	if (verbose)
+	{
+		if (verbose)
+			cat(paste("--- Compute ",K," synchrones with ",nb_ref_curves," series\n", sep=""))
+	}
 	indices_workers = .spreadIndices(seq_len(nb_ref_curves), nb_series_per_chunk)
 	ignored <-
 		if (parll)
-			parallel::parLapply(indices_workers, computeSynchronesChunk)
+			parallel::parLapply(cl, indices_workers, computeSynchronesChunk)
 		else
 			lapply(indices_workers, computeSynchronesChunk)
 
 	if (parll)
 		parallel::stopCluster(cl)
 
-	#TODO: can we avoid this loop? ( synchrones = sweep(synchrones, 1, counts, '/') )
+	if (!sync_mean)
+		return (synchrones)
+
+	#TODO: can we avoid this loop? ( synchrones = sweep(synchrones, 2, counts, '/') )
 	for (i in seq_len(K))
-		synchrones[i,] = synchrones[i,] / counts[i,1]
+		synchrones[,i] = synchrones[,i] / counts[i]
 	#NOTE: odds for some clusters to be empty? (when series already come from stage 2)
 	#      ...maybe; but let's hope resulting K1' be still quite bigger than K2
-	noNA_rows = sapply(seq_len(K), function(i) all(!is.nan(synchrones[i,])))
+	noNA_rows = sapply(seq_len(K), function(i) all(!is.nan(synchrones[,i])))
 	if (all(noNA_rows))
 		return (synchrones)
 	# Else: some clusters are empty, need to slice synchrones
-	synchrones[noNA_rows,]
+	bigmemory::as.big.matrix(synchrones[,noNA_rows])
 }
 
 #' computeWerDists
@@ -171,17 +191,11 @@ computeSynchrones = function(medoids, getRefSeries,
 #'   as the series in the initial dataset
 #' @inheritParams claws
 #'
-#' @return A big.matrix of size K1 x K1
+#' @return A matrix of size K1 x K1
 #'
 #' @export
-computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE)
+computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALSE,parll=TRUE)
 {
-
-
-
-#TODO: re-organize to call computeWerDist(x,y) [C] (in //?) from two indices + big.matrix
-
-
 	n <- nrow(synchrones)
 	delta <- ncol(synchrones)
 	#TODO: automatic tune of all these parameters ? (for other users)
@@ -192,106 +206,130 @@ computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE)
 	#NOTE: default scalevector == 2^(0:(noctave * nvoice) / nvoice) * s0 (?)
 	scalevector  <- 2^(4:(noctave * nvoice) / nvoice + 1)
 	#condition: ( log2(s0*w0/(2*pi)) - 1 ) * nvoice + 1.5 >= 1
-	s0=2
-	w0=2*pi
+	s0 = 2
+	w0 = 2*pi
 	scaled=FALSE
 	s0log = as.integer( (log2( s0*w0/(2*pi) ) - 1) * nvoice + 1.5 )
 	totnoct = noctave + as.integer(s0log/nvoice) + 1
 
-	computeCWT = function(i)
+	Xwer_dist <- bigmemory::big.matrix(nrow=n, ncol=n, type="double")
+
+	cwt_file = ".epclust_bin/cwt"
+	#TODO: args, nb_per_chunk, nbytes, endian
+
+	# Generate n(n-1)/2 pairs for WER distances computations
+	pairs = list()
+	V = seq_len(n)
+	for (i in 1:n)
 	{
-		if (verbose)
-			cat(paste("+++ Compute Rwave::cwt() on serie ",i,"\n", sep=""))
-		ts <- scale(ts(synchrones[i,]), center=TRUE, scale=scaled)
+		V = V[-1]
+		pairs = c(pairs, lapply(V, function(v) c(i,v)))
+	}
+
+	computeSaveCWT = function(index)
+	{
+		ts <- scale(ts(synchrones[index,]), center=TRUE, scale=scaled)
 		totts.cwt = Rwave::cwt(ts, totnoct, nvoice, w0, plot=FALSE)
 		ts.cwt = totts.cwt[,s0log:(s0log+noctave*nvoice)]
 		#Normalization
 		sqs <- sqrt(2^(0:(noctave*nvoice)/nvoice)*s0)
 		sqres <- sweep(ts.cwt,2,sqs,'*')
-		sqres / max(Mod(sqres))
+		res <- sqres / max(Mod(sqres))
+		#TODO: serializer les CWT, les récupérer via getDataInFile ;
+		#--> OK, faut juste stocker comme séries simples de taille delta*ncol (53*17519)
+		binarize(c(as.double(Re(res)),as.double(Im(res))), cwt_file, ncol(res), ",", nbytes, endian)
 	}
 
 	if (parll)
 	{
 		cl = parallel::makeCluster(ncores_clust)
-		parallel::clusterExport(cl,
-			varlist=c("synchrones","totnoct","nvoice","w0","s0log","noctave","s0","verbose"),
-			envir=environment())
+		synchrones_desc <- bigmemory::describe(synchrones)
+		Xwer_dist_desc <- bigmemory::describe(Xwer_dist)
+		parallel::clusterExport(cl, varlist=c("synchrones_desc","Xwer_dist_desc","totnoct",
+			"nvoice","w0","s0log","noctave","s0","verbose","getCWT"), envir=environment())
 	}
-
-	# list of CWT from synchrones
-	# TODO: fit in RAM, OK? If not, 2 options: serialize, compute individual distances
-	Xcwt4 <-
+	
+	if (verbose)
+	{
+		cat(paste("--- Compute WER dists\n", sep=""))
+	#	precompute save all CWT........
+	}
+	#precompute and serialize all CWT
+	ignored <-
 		if (parll)
-			parallel::parLapply(cl, seq_len(n), computeCWT)
+			parallel::parLapply(cl, 1:n, computeSaveCWT)
 		else
-			lapply(seq_len(n), computeCWT)
-
-	if (parll)
-		parallel::stopCluster(cl)
+			lapply(1:n, computeSaveCWT)
 
-	Xwer_dist <- bigmemory::big.matrix(nrow=n, ncol=n, type="double")
-	fcoefs = rep(1/3, 3) #moving average on 3 values (TODO: very slow! correct?!)
-	if (verbose)
-		cat("*** Compute WER distances from CWT\n")
-
-	#TODO: computeDistances(i,j), et répartir les n(n-1)/2 couples d'indices
-	#là c'est trop déséquilibré
+	getCWT = function(index)
+	{
+		#from cwt_file ...
+		res <- getDataInFile(c(2*index-1,2*index), cwt_file, nbytes, endian)
+	###############TODO:
+	}
 
-	computeDistancesLineI = function(i)
+	# Distance between rows i and j
+	computeDistancesIJ = function(pair)
 	{
-		if (verbose)
-			cat(paste("   Line ",i,"\n", sep=""))
-		for (j in (i+1):n)
+		if (parll)
 		{
-			#TODO: 'circular=TRUE' is wrong, should just take values on the sides; to rewrite in C
-			num <- filter(Mod(Xcwt4[[i]] * Conj(Xcwt4[[j]])), fcoefs, circular=TRUE)
-			WX <- filter(Mod(Xcwt4[[i]] * Conj(Xcwt4[[i]])), fcoefs, circular=TRUE)
-			WY <- filter(Mod(Xcwt4[[j]] * Conj(Xcwt4[[j]])), fcoefs, circular=TRUE)
-			wer2 <- sum(colSums(num)^2) / sum( sum(colSums(WX) * colSums(WY)) )
-			if (parll)
-				synchronicity::lock(m)
-			Xwer_dist[i,j] <- sqrt(delta * ncol(Xcwt4[[1]]) * (1 - wer2))
-			Xwer_dist[j,i] <- Xwer_dist[i,j]
-			if (parll)
-				synchronicity::unlock(m)
+			require("bigmemory", quietly=TRUE)
+			require("epclust", quietly=TRUE)
+			synchrones <- bigmemory::attach.big.matrix(synchrones_desc)
+			Xwer_dist <- bigmemory::attach.big.matrix(Xwer_dist_desc)
 		}
+
+		i = pair[1] ; j = pair[2]
+		if (verbose && j==i+1)
+			cat(paste("   Distances (",i,",",j,"), (",i,",",j+1,") ...\n", sep=""))
+		cwt_i <- getCWT(i)
+		cwt_j <- getCWT(j)
+
+		num <- epclustFilter(Mod(cwt_i * Conj(cwt_j)))
+		WX  <- epclustFilter(Mod(cwt_i * Conj(cwt_i)))
+		WY <- epclustFilter(Mod(cwt_j * Conj(cwt_j)))
+		wer2 <- sum(colSums(num)^2) / sum(colSums(WX) * colSums(WY))
+		Xwer_dist[i,j] <- sqrt(delta * ncol(cwt_i) * max(1 - wer2, 0.)) #FIXME: wer2 should be < 1
+		Xwer_dist[j,i] <- Xwer_dist[i,j]
 		Xwer_dist[i,i] = 0.
 	}
 
-	parll = (requireNamespace("synchronicity",quietly=TRUE)
-		&& parll && Sys.info()['sysname'] != "Windows")
-	if (parll)
-		m <- synchronicity::boost.mutex()
-
+	if (verbose)
+	{
+		cat(paste("--- Compute WER dists\n", sep=""))
+	}
 	ignored <-
 		if (parll)
-		{
-			parallel::mclapply(seq_len(n-1), computeDistancesLineI,
-				mc.cores=ncores_clust, mc.allow.recursive=FALSE)
-		}
+			parallel::parLapply(cl, pairs, computeDistancesIJ)
 		else
-			lapply(seq_len(n-1), computeDistancesLineI)
+			lapply(pairs, computeDistancesIJ)
+
+	if (parll)
+		parallel::stopCluster(cl)
+
 	Xwer_dist[n,n] = 0.
-	Xwer_dist
+	distances <- Xwer_dist[,]
+	rm(Xwer_dist) ; gc()
+	distances #~small matrix K1 x K1
 }
 
 # Helper function to divide indices into balanced sets
-.spreadIndices = function(indices, nb_per_chunk)
+.spreadIndices = function(indices, nb_per_set)
 {
 	L = length(indices)
-	nb_workers = floor( L / nb_per_chunk )
-	if (nb_workers == 0)
+	nb_workers = floor( L / nb_per_set )
+	rem = L %% nb_per_set
+	if (nb_workers == 0 || (nb_workers==1 && rem==0))
 	{
-		# L < nb_series_per_chunk, simple case
+		# L <= nb_per_set, simple case
 		indices_workers = list(indices)
 	}
 	else
 	{
 		indices_workers = lapply( seq_len(nb_workers), function(i)
-			indices[(nb_per_chunk*(i-1)+1):(nb_per_chunk*i)] )
+			indices[(nb_per_set*(i-1)+1):(nb_per_set*i)] )
 		# Spread the remaining load among the workers
-		rem = L %% nb_per_chunk
+		rem = L %% nb_per_set
 		while (rem > 0)
 		{
 			index = rem%%nb_workers + 1