X-Git-Url: https://git.auder.net/?p=epclust.git;a=blobdiff_plain;f=epclust%2FR%2Fclustering.R;h=2ce4267ef35e717a7bea0a8667f648367d38fbcf;hp=a431ba86e6120aef106fdce8712eb52a5ef3c4b0;hb=d9bb53c5e1392018bf67f92140edb10137f3423c;hpb=9f05a4a0b703deffd7bdb9cd99b0aaa2246a5c83 diff --git a/epclust/R/clustering.R b/epclust/R/clustering.R index a431ba8..2ce4267 100644 --- a/epclust/R/clustering.R +++ b/epclust/R/clustering.R @@ -33,10 +33,12 @@ clusteringTask1 = function(indices, getContribs, K1, algoClust1, nb_items_clust1 if (parll) { cl = parallel::makeCluster(ncores_clust, outfile = "") - parallel::clusterExport(cl, varlist=c("getContribs","K1","verbose"), envir=environment()) + 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 = .spreadIndices(indices, nb_items_clust1) if (verbose) cat(paste("*** [iterated] Clustering task 1 on ",length(indices)," series\n", sep="")) @@ -64,16 +66,23 @@ clusteringTask1 = function(indices, getContribs, K1, algoClust1, nb_items_clust1 #' @rdname clustering #' @export clusteringTask2 = function(medoids, K2, algoClust2, getRefSeries, nb_ref_curves, - nb_series_per_chunk, sync_mean, nbytes,endian,ncores_clust=1,verbose=FALSE,parll=TRUE) + nb_series_per_chunk, nbytes,endian,ncores_clust=1,verbose=FALSE,parll=TRUE) { if (verbose) cat(paste("*** Clustering task 2 on ",ncol(medoids)," synchrones\n", sep="")) if (ncol(medoids) <= K2) return (medoids) + + # A) Obtain synchrones, that is to say the cumulated power consumptions + # for each of the K1 initial groups synchrones = computeSynchrones(medoids, getRefSeries, nb_ref_curves, - nb_series_per_chunk, sync_mean, ncores_clust, verbose, parll) + nb_series_per_chunk, ncores_clust, verbose, parll) + + # B) Compute the WER distances (Wavelets Extended coefficient of deteRmination) distances = computeWerDists(synchrones, nbytes, endian, ncores_clust, verbose, parll) + + # C) Apply clustering algorithm 2 on the WER distances matrix if (verbose) cat(paste(" algoClust2() on ",nrow(distances)," items\n", sep="")) medoids[ ,algoClust2(distances,K2) ] @@ -82,7 +91,7 @@ clusteringTask2 = function(medoids, K2, algoClust2, getRefSeries, nb_ref_curves, #' computeSynchrones #' #' Compute the synchrones curves (sum of clusters elements) from a matrix of medoids, -#' using L2 distances. +#' using euclidian distance. #' #' @param medoids big.matrix of medoids (curves of same length as initial series) #' @param getRefSeries Function to retrieve initial series (e.g. in stage 2 after series @@ -94,8 +103,9 @@ clusteringTask2 = function(medoids, K2, algoClust2, getRefSeries, nb_ref_curves, #' #' @export computeSynchrones = function(medoids, getRefSeries, nb_ref_curves, - nb_series_per_chunk, sync_mean, ncores_clust=1,verbose=FALSE,parll=TRUE) + nb_series_per_chunk, ncores_clust=1,verbose=FALSE,parll=TRUE) { + # Synchrones computation is embarassingly parallel: compute it by chunks of series computeSynchronesChunk = function(indices) { if (parll) @@ -103,26 +113,25 @@ computeSynchrones = function(medoids, getRefSeries, nb_ref_curves, require("bigmemory", quietly=TRUE) requireNamespace("synchronicity", quietly=TRUE) require("epclust", quietly=TRUE) + # The big.matrix objects need to be attached to be usable on the workers 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) } + # Obtain a chunk of reference series ref_series = getRefSeries(indices) nb_series = ncol(ref_series) # Get medoids indices for this chunk of series mi = computeMedoidsIndices(medoids@address, ref_series) + # Update synchrones using mi above for (i in seq_len(nb_series)) { if (parll) - synchronicity::lock(m) + synchronicity::lock(m) #locking required because several writes at the same time synchrones[, mi[i] ] = synchrones[, mi[i] ] + ref_series[,i] - if (sync_mean) - counts[ mi[i] ] = counts[ mi[i] ] + 1 if (parll) synchronicity::unlock(m) } @@ -130,34 +139,29 @@ computeSynchrones = function(medoids, getRefSeries, nb_ref_curves, 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=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 + # NOTE: 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() + m <- synchronicity::boost.mutex() #for lock/unlock, see computeSynchronesChunk + # mutex and big.matrix objects cannot be passed directly: + # they will be accessed from their description 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) - 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()) + parallel::clusterExport(cl, envir=environment(), + varlist=c("synchrones_desc","m_desc","medoids_desc","getRefSeries")) } 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) + cat(paste("--- Compute ",K," synchrones with ",nb_ref_curves," series\n", sep="")) + + # Balance tasks by splitting the indices set - maybe not so evenly, but + # max==TRUE in next call ensures that no set has more than nb_series_per_chunk items. + indices_workers = .spreadIndices(seq_len(nb_ref_curves), nb_series_per_chunk, max=TRUE) ignored <- if (parll) parallel::parLapply(cl, indices_workers, computeSynchronesChunk) @@ -167,19 +171,7 @@ computeSynchrones = function(medoids, getRefSeries, nb_ref_curves, if (parll) parallel::stopCluster(cl) - 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] - #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]))) - if (all(noNA_rows)) - return (synchrones) - # Else: some clusters are empty, need to slice synchrones - bigmemory::as.big.matrix(synchrones[,noNA_rows]) + return (synchrones) } #' computeWerDists @@ -196,21 +188,13 @@ computeSynchrones = function(medoids, getRefSeries, nb_ref_curves, #' @export computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALSE,parll=TRUE) { - n <- nrow(synchrones) - delta <- ncol(synchrones) + n <- ncol(synchrones) + L <- nrow(synchrones) #TODO: automatic tune of all these parameters ? (for other users) + # 4 here represent 2^5 = 32 half-hours ~ 1 day nvoice <- 4 # noctave = 2^13 = 8192 half hours ~ 180 days ; ~log2(ncol(synchrones)) 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 + 1) - #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 Xwer_dist <- bigmemory::big.matrix(nrow=n, ncol=n, type="double") @@ -228,15 +212,15 @@ computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALS computeSaveCWT = function(index) { - ts <- scale(ts(synchrones[index,]), center=TRUE, scale=scaled) - totts.cwt = Rwave::cwt(ts, totnoct, nvoice, w0, plot=FALSE) + ts <- scale(ts(synchrones[,index]), center=TRUE, scale=FALSE) + totts.cwt = Rwave::cwt(ts, totnoct, nvoice, w0=2*pi, twoD=TRUE, 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,'*') 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) + #--> OK, faut juste stocker comme séries simples de taille L*n' (53*17519) binarize(c(as.double(Re(res)),as.double(Im(res))), cwt_file, ncol(res), ",", nbytes, endian) } @@ -245,8 +229,9 @@ computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALS cl = parallel::makeCluster(ncores_clust) 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()) + parallel::clusterExport(cl, envir=environment(), + varlist=c("synchrones_desc","Xwer_dist_desc","totnoct","nvoice","w0","s0log", + "noctave","s0","verbose","getCWT")) } if (verbose) @@ -289,7 +274,7 @@ computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALS 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[i,j] <- sqrt(L * ncol(cwt_i) * max(1 - wer2, 0.)) Xwer_dist[j,i] <- Xwer_dist[i,j] Xwer_dist[i,i] = 0. } @@ -314,7 +299,8 @@ computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALS } # Helper function to divide indices into balanced sets -.spreadIndices = function(indices, nb_per_set) +# If max == TRUE, sets sizes cannot exceed nb_per_set +.spreadIndices = function(indices, nb_per_set, max=FALSE) { L = length(indices) nb_workers = floor( L / nb_per_set ) @@ -328,6 +314,13 @@ computeWerDists = function(synchrones, nbytes,endian,ncores_clust=1,verbose=FALS { indices_workers = lapply( seq_len(nb_workers), function(i) indices[(nb_per_set*(i-1)+1):(nb_per_set*i)] ) + + if (max) + { + # Sets are not so well balanced, but size is supposed to be critical + return ( c( indices_workers, (L-rem+1):L ) ) + } + # Spread the remaining load among the workers rem = L %% nb_per_set while (rem > 0)