X-Git-Url: https://git.auder.net/?a=blobdiff_plain;f=epclust%2FR%2Fclustering.R;h=5b5f6684763321b1b854d0cdbc2de6bb2b8ded16;hb=074a48c472fcbdf99a36fae333dd8dbb568c06a0;hp=fce1b1c695f4b6094580375e28a6cd5cd0f1e805;hpb=56857861dc15088cf58e7438968fe5714b22168e;p=epclust.git

diff --git a/epclust/R/clustering.R b/epclust/R/clustering.R
index fce1b1c..5b5f668 100644
--- a/epclust/R/clustering.R
+++ b/epclust/R/clustering.R
@@ -1,111 +1,94 @@
-# Cluster one full task (nb_curves / ntasks series); only step 1
-clusteringTask = function(indices, getCoefs, K1, nb_series_per_chunk, ncores)
-{
-	cl = parallel::makeCluster(ncores)
-	repeat
-	{
-		nb_workers = max( 1, round( length(indices) / nb_series_per_chunk ) )
-		indices_workers = lapply(seq_len(nb_workers), function(i) {
-			upper_bound = ifelse( i<nb_workers,
-				min(nb_series_per_chunk*i,length(indices)), length(indices) )
-			indices[(nb_series_per_chunk*(i-1)+1):upper_bound]
-		})
-		indices = unlist( parallel::parLapply(cl, indices_workers, function(inds)
-			computeClusters1(getCoefs(inds), K1)) )
-		if (length(indices) == K1)
-			break
-	}
-	parallel::stopCluster(cl)
-	indices #medoids
-}
+#' Two-stage clustering, within one task (see \code{claws()})
+#'
+#' \code{clusteringTask1()} runs one full stage-1 task, which consists in iterated
+#' clustering on nb_curves / ntasks energy contributions, computed through
+#' discrete wavelets coefficients.
+#' \code{clusteringTask2()} runs a full stage-2 task, which consists in WER distances
+#' computations between medoids (indices) output from stage 1, before applying
+#' the second clustering algorithm on the distances matrix.
+#'
+#' @param getContribs Function to retrieve contributions from initial series indices:
+#'   \code{getContribs(indices)} outputs a contributions matrix, in columns
+#' @inheritParams claws
+#' @inheritParams computeSynchrones
+#' @inheritParams computeWerDists
+#'
+#' @return The indices of the computed (resp. K1 and K2) medoids.
+#'
+#' @name clustering
+#' @rdname clustering
+#' @aliases clusteringTask1 clusteringTask2
+NULL
 
-# Apply the clustering algorithm (PAM) on a coeffs or distances matrix
-computeClusters1 = function(coefs, K1)
-	indices[ cluster::pam(coefs, K1, diss=FALSE)$id.med ]
-
-# Cluster a chunk of series inside one task (~max nb_series_per_chunk)
-computeClusters2 = function(medoids, K2, getRefSeries, nb_series_per_chunk)
+#' @rdname clustering
+#' @export
+clusteringTask1 <- function(indices, getContribs, K1, algoClust1, nb_items_clust,
+	ncores_clust=3, verbose=FALSE)
 {
-	synchrones = computeSynchrones(medoids, getRefSeries, nb_series_per_chunk)
-	cluster::pam(computeWerDists(synchrones), K2, diss=TRUE)$medoids
-}
+	if (verbose)
+		cat(paste("*** Clustering task 1 on ",length(indices)," series [start]\n", sep=""))
 
-# Compute the synchrones curves (sum of clusters elements) from a clustering result
-computeSynchrones = function(medoids, getRefSeries, nb_series_per_chunk)
-{
-	#les getSeries(indices) sont les medoides --> init vect nul pour chacun, puis incr avec les
-	#courbes (getSeriesForSynchrones) les plus proches... --> au sens de la norme L2 ?
-	K = nrow(medoids)
-	synchrones = matrix(0, nrow=K, ncol=ncol(medoids))
-	counts = rep(0,K)
-	index = 1
-	repeat
+	if (length(indices) <= K1)
+		return (indices)
+
+	parll <- (ncores_clust > 1)
+	if (parll)
 	{
-		range = (index-1) + seq_len(nb_series_per_chunk)
-		ref_series = getRefSeries(range)
-		if (is.null(ref_series))
-			break
-		#get medoids indices for this chunk of series
-		for (i in seq_len(nrow(ref_series)))
+		# outfile=="" to see stderr/stdout on terminal
+		cl <-
+			if (verbose)
+				parallel::makeCluster(ncores_clust, outfile = "")
+			else
+				parallel::makeCluster(ncores_clust)
+		parallel::clusterExport(cl, c("getContribs","K1","verbose"), envir=environment())
+	}
+	# Iterate clustering algorithm 1 until K1 medoids are found
+	while (length(indices) > K1)
+	{
+		# Balance tasks by splitting the indices set - as evenly as possible
+		indices_workers <- .splitIndices(indices, nb_items_clust, min_size=K1+1)
+		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 (verbose)
 		{
-			j = which.min( rowSums( sweep(medoids, 2, series[i,], '-')^2 ) )
-			synchrones[j,] = synchrones[j,] + series[i,]
-			counts[j] = counts[j] + 1
+			cat(paste("*** Clustering task 1 on ",length(indices)," medoids [iter]\n", sep=""))
 		}
-		index = index + nb_series_per_chunk
 	}
-	#NOTE: odds for some clusters to be empty? (when series already come from stage 2)
-	sweep(synchrones, 1, counts, '/')
+	if (parll)
+		parallel::stopCluster(cl)
+
+	indices #medoids
 }
 
-# Compute the WER distance between the synchrones curves (in rows)
-computeWerDist = function(curves)
+#' @rdname clustering
+#' @export
+clusteringTask2 <- function(indices, getSeries, K2, algoClust2, nb_series_per_chunk,
+	smooth_lvl, nvoice, nbytes, endian, ncores_clust=3, verbose=FALSE)
 {
-	if (!require("Rwave", quietly=TRUE))
-		stop("Unable to load Rwave library")
-	n <- nrow(curves)
-	delta <- ncol(curves)
-	#TODO: automatic tune of all these parameters ? (for other users)
-	nvoice   <- 4
-	# noctave = 2^13 = 8192 half hours ~ 180 days ; ~log2(ncol(curves))
-	noctave = 13
-	# 4 here represent 2^5 = 32 half-hours ~ 1 day
-	#NOTE: default scalevector == 2^(0:(noctave * nvoice) / nvoice) * s0 (?)
-	scalevector  <- 2^(4:(noctave * nvoice) / nvoice) * 2
-	#condition: ( log2(s0*w0/(2*pi)) - 1 ) * nvoice + 1.5 >= 1
-	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
+	if (verbose)
+		cat(paste("*** Clustering task 2 on ",length(indices)," medoids\n", sep=""))
 
-	# (normalized) observations node with CWT
-	Xcwt4 <- lapply(seq_len(n), function(i) {
-		ts <- scale(ts(curves[i,]), center=TRUE, scale=scaled)
-		totts.cwt = Rwave::cwt(ts,totnoct,nvoice,w0,plot=0)
-		ts.cwt = totts.cwt[,s0log:(s0log+noctave*nvoice)]
-		#Normalization
-		sqs <- sqrt(2^(0:(noctave*nvoice)/nvoice)*s0)
-		sqres <- sweep(ts.cwt,MARGIN=2,sqs,'*')
-		sqres / max(Mod(sqres))
-	})
+	if (length(indices) <= K2)
+		return (indices)
 
-	Xwer_dist <- matrix(0., n, n)
-	fcoefs = rep(1/3, 3) #moving average on 3 values (TODO: very slow! correct?!)
-	for (i in 1:(n-1))
-	{
-		for (j in (i+1):n)
-		{
-			#TODO: later, compute CWT here (because not enough storage space for 200k series)
-			#      '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)) )
-			Xwer_dist[i,j] <- sqrt(delta * ncol(Xcwt4[[1]]) * (1 - wer2))
-			Xwer_dist[j,i] <- Xwer_dist[i,j]
-		}
-	}
-	diag(Xwer_dist) <- numeric(n)
-	Xwer_dist
+	# A) Compute the WER distances (Wavelets Extended coefficient of deteRmination)
+	distances <- computeWerDists(indices, getSeries, nb_series_per_chunk,
+		smooth_lvl, nvoice, nbytes, endian, ncores_clust, verbose)
+
+	# B) Apply clustering algorithm 2 on the WER distances matrix
+	if (verbose)
+		cat(paste("*** algoClust2() on ",nrow(distances)," items\n", sep=""))
+	indices[ algoClust2(distances,K2) ]
 }