epclust/R/clustering.R

   1 #' @name clustering
   2 #' @rdname clustering
   3 #' @aliases clusteringTask1 computeClusters1 computeClusters2
   4 #'
   5 #' @title Two-stage clustering, withing one task (see \code{claws()})
   6 #'
   7 #' @description \code{clusteringTask1()} runs one full stage-1 task, which consists in
   8 #'   iterated stage 1 clustering (on nb_curves / ntasks energy contributions, computed
   9 #'   through discrete wavelets coefficients).
  10 #'   \code{clusteringTask2()} runs a full stage-2 task, which consists in synchrones
  11 #'   and then WER distances computations, before applying the clustering algorithm.
  12 #'   \code{computeClusters1()} and \code{computeClusters2()} correspond to the atomic
  13 #'   clustering procedures respectively for stage 1 and 2. The former applies the
  14 #'   clustering algorithm (PAM) on a contributions matrix, while the latter clusters
  15 #'   a chunk of series inside one task (~max nb_series_per_chunk)
  16 #'
  17 #' @param indices Range of series indices to cluster in parallel (initial data)
  18 #' @param getContribs Function to retrieve contributions from initial series indices:
  19 #'   \code{getContribs(indices)} outpus a contributions matrix
  20 #' @param contribs matrix of contributions (e.g. output of \code{curvesToContribs()})
  21 #' @param distances matrix of K1 x K1 (WER) distances between synchrones
  22 #' @inheritParams computeSynchrones
  23 #' @inheritParams claws
  24 #'
  25 #' @return For \code{clusteringTask1()} and \code{computeClusters1()}, the indices of the
  26 #'   computed (K1) medoids. Indices are irrelevant for stage 2 clustering, thus
  27 #'   \code{computeClusters2()} outputs a big.matrix of medoids
  28 #'   (of size limited by nb_series_per_chunk)
  29 NULL
  30
  31 #' @rdname clustering
  32 #' @export
  33 clusteringTask1 = function(
  34     indices, getContribs, K1, nb_series_per_chunk, ncores_clust=1, verbose=FALSE, parll=TRUE)
  35 {
  36     if (verbose)
  37         cat(paste("*** Clustering task 1 on ",length(indices)," lines\n", sep=""))
  38
  39     if (parll)
  40     {
  41         cl = parallel::makeCluster(ncores_clust)
  42         parallel::clusterExport(cl, varlist=c("getContribs","K1","verbose"), envir=environment())
  43     }
  44     while (length(indices) > K1)
  45     {
  46         indices_workers = .spreadIndices(indices, nb_series_per_chunk)
  47         indices <-
  48             if (parll)
  49             {
  50                 unlist( parallel::parLapply(cl, indices_workers, function(inds) {
  51                     require("epclust", quietly=TRUE)
  52                     inds[ computeClusters1(getContribs(inds), K1, verbose) ]
  53                 }) )
  54             }
  55             else
  56             {
  57                 unlist( lapply(indices_workers, function(inds)
  58                     inds[ computeClusters1(getContribs(inds), K1, verbose) ]
  59                 ) )
  60             }
  61     }
  62     if (parll)
  63         parallel::stopCluster(cl)
  64
  65     indices #medoids
  66 }
  67
  68 #' @rdname clustering
  69 #' @export
  70 clusteringTask2 = function(medoids, K2,
  71     getRefSeries, nb_ref_curves, nb_series_per_chunk, ncores_clust=1,verbose=FALSE,parll=TRUE)
  72 {
  73     if (verbose)
  74         cat(paste("*** Clustering task 2 on ",nrow(medoids)," lines\n", sep=""))
  75
  76     if (nrow(medoids) <= K2)
  77         return (medoids)
  78     synchrones = computeSynchrones(medoids,
  79         getRefSeries, nb_ref_curves, nb_series_per_chunk, ncores_clust, verbose, parll)
  80     distances = computeWerDists(synchrones, ncores_clust, verbose, parll)
  81     medoids[ computeClusters2(distances,K2,verbose), ]
  82 }
  83
  84 #' @rdname clustering
  85 #' @export
  86 computeClusters1 = function(contribs, K1, verbose=FALSE)
  87 {
  88     if (verbose)
  89         cat(paste("   computeClusters1() on ",nrow(contribs)," lines\n", sep=""))
  90     cluster::pam(contribs, K1, diss=FALSE)$id.med
  91 }
  92
  93 #' @rdname clustering
  94 #' @export
  95 computeClusters2 = function(distances, K2, verbose=FALSE)
  96 {
  97     if (verbose)
  98         cat(paste("   computeClusters2() on ",nrow(distances)," lines\n", sep=""))
  99     cluster::pam(distances, K2, diss=TRUE)$id.med
 100 }
 101
 102 #' computeSynchrones
 103 #'
 104 #' Compute the synchrones curves (sum of clusters elements) from a matrix of medoids,
 105 #' using L2 distances.
 106 #'
 107 #' @param medoids big.matrix of medoids (curves of same length as initial series)
 108 #' @param getRefSeries Function to retrieve initial series (e.g. in stage 2 after series
 109 #'   have been replaced by stage-1 medoids)
 110 #' @param nb_ref_curves How many reference series? (This number is known at this stage)
 111 #' @inheritParams claws
 112 #'
 113 #' @return A big.matrix of size K1 x L where L = data_length
 114 #'
 115 #' @export
 116 computeSynchrones = function(medoids, getRefSeries,
 117     nb_ref_curves, nb_series_per_chunk, ncores_clust=1,verbose=FALSE,parll=TRUE)
 118 {
 119     if (verbose)
 120         cat(paste("--- Compute synchrones\n", sep=""))
 121
 122     computeSynchronesChunk = function(indices)
 123     {
 124         if (parll)
 125         {
 126             require("bigmemory", quietly=TRUE)
 127             requireNamespace("synchronicity", quietly=TRUE)
 128             require("epclust", quietly=TRUE)
 129             synchrones <- bigmemory::attach.big.matrix(synchrones_desc)
 130             counts <- bigmemory::attach.big.matrix(counts_desc)
 131             medoids <- bigmemory::attach.big.matrix(medoids_desc)
 132             m <- synchronicity::attach.mutex(m_desc)
 133         }
 134
 135         ref_series = getRefSeries(indices)
 136         nb_series = nrow(ref_series)
 137
 138         #get medoids indices for this chunk of series
 139         mi = computeMedoidsIndices(medoids@address, ref_series)
 140
 141         for (i in seq_len(nb_series))
 142         {
 143             if (parll)
 144                 synchronicity::lock(m)
 145             synchrones[ mi[i], ] = synchrones[ mi[i], ] + ref_series[i,]
 146             counts[ mi[i] ] = counts[ mi[i] ] + 1 #TODO: remove counts?
 147             if (parll)
 148                 synchronicity::unlock(m)
 149         }
 150     }
 151
 152     K = nrow(medoids) ; L = ncol(medoids)
 153     # Use bigmemory (shared==TRUE by default) + synchronicity to fill synchrones in //
 154     # TODO: if size > RAM (not our case), use file-backed big.matrix
 155     synchrones = bigmemory::big.matrix(nrow=K, ncol=L, type="double", init=0.)
 156     counts = bigmemory::big.matrix(nrow=K, ncol=1, type="double", init=0)
 157     # synchronicity is only for Linux & MacOS; on Windows: run sequentially
 158     parll = (requireNamespace("synchronicity",quietly=TRUE)
 159         && parll && Sys.info()['sysname'] != "Windows")
 160     if (parll)
 161     {
 162         m <- synchronicity::boost.mutex()
 163         m_desc <- synchronicity::describe(m)
 164         synchrones_desc = bigmemory::describe(synchrones)
 165         counts_desc = bigmemory::describe(counts)
 166         medoids_desc = bigmemory::describe(medoids)
 167         cl = parallel::makeCluster(ncores_clust)
 168         parallel::clusterExport(cl, varlist=c("synchrones_desc","counts_desc","counts",
 169             "verbose","m_desc","medoids_desc","getRefSeries"), envir=environment())
 170     }
 171
 172     indices_workers = .spreadIndices(seq_len(nb_ref_curves), nb_series_per_chunk)
 173     ignored <-
 174         if (parll)
 175             parallel::parLapply(cl, indices_workers, computeSynchronesChunk)
 176         else
 177             lapply(indices_workers, computeSynchronesChunk)
 178
 179     if (parll)
 180         parallel::stopCluster(cl)
 181
 182     #TODO: can we avoid this loop? ( synchrones = sweep(synchrones, 1, counts, '/') )
 183     for (i in seq_len(K))
 184         synchrones[i,] = synchrones[i,] / counts[i,1]
 185     #NOTE: odds for some clusters to be empty? (when series already come from stage 2)
 186     #      ...maybe; but let's hope resulting K1' be still quite bigger than K2
 187     noNA_rows = sapply(seq_len(K), function(i) all(!is.nan(synchrones[i,])))
 188     if (all(noNA_rows))
 189         return (synchrones)
 190     # Else: some clusters are empty, need to slice synchrones
 191     synchrones[noNA_rows,]
 192 }
 193
 194 #' computeWerDists
 195 #'
 196 #' Compute the WER distances between the synchrones curves (in rows), which are
 197 #' returned (e.g.) by \code{computeSynchrones()}
 198 #'
 199 #' @param synchrones A big.matrix of synchrones, in rows. The series have same length
 200 #'   as the series in the initial dataset
 201 #' @inheritParams claws
 202 #'
 203 #' @return A matrix of size K1 x K1
 204 #'
 205 #' @export
 206 computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE)
 207 {
 208     if (verbose)
 209         cat(paste("--- Compute WER dists\n", sep=""))
 210
 211
 212
 213
 214 #TODO: serializer les CWT, les récupérer via getDataInFile
 215 #--> OK, faut juste stocker comme séries simples de taille delta*ncol (53*17519)
 216
 217
 218
 219
 220     n <- nrow(synchrones)
 221     delta <- ncol(synchrones)
 222     #TODO: automatic tune of all these parameters ? (for other users)
 223     nvoice   <- 4
 224     # noctave = 2^13 = 8192 half hours ~ 180 days ; ~log2(ncol(synchrones))
 225     noctave = 13
 226     # 4 here represent 2^5 = 32 half-hours ~ 1 day
 227     #NOTE: default scalevector == 2^(0:(noctave * nvoice) / nvoice) * s0 (?)
 228     scalevector  <- 2^(4:(noctave * nvoice) / nvoice + 1)
 229     #condition: ( log2(s0*w0/(2*pi)) - 1 ) * nvoice + 1.5 >= 1
 230     s0=2
 231     w0=2*pi
 232     scaled=FALSE
 233     s0log = as.integer( (log2( s0*w0/(2*pi) ) - 1) * nvoice + 1.5 )
 234     totnoct = noctave + as.integer(s0log/nvoice) + 1
 235
 236     Xwer_dist <- bigmemory::big.matrix(nrow=n, ncol=n, type="double")
 237
 238     # Generate n(n-1)/2 pairs for WER distances computations
 239 #   pairs = list()
 240 #   V = seq_len(n)
 241 #   for (i in 1:n)
 242 #   {
 243 #       V = V[-1]
 244 #       pairs = c(pairs, lapply(V, function(v) c(i,v)))
 245 #   }
 246     # Generate "smart" pairs for WER distances computations
 247     pairs = list()
 248     F = floor(2*n/3)
 249     for (i in 1:F)
 250         pairs = c(pairs, lapply((i+1):n, function(v) c(i,v)))
 251     V = (F+1):n
 252     for (i in (F+1):(n-1))
 253     {
 254         V = V[-1]
 255         pairs = c(pairs,
 256
 257     # Distance between rows i and j
 258     computeDistancesIJ = function(pair)
 259     {
 260         if (parll)
 261         {
 262             require("bigmemory", quietly=TRUE)
 263             require("epclust", quietly=TRUE)
 264             synchrones <- bigmemory::attach.big.matrix(synchrones_desc)
 265             Xwer_dist <- bigmemory::attach.big.matrix(Xwer_dist_desc)
 266         }
 267
 268         computeCWT = function(index)
 269         {
 270             ts <- scale(ts(synchrones[index,]), center=TRUE, scale=scaled)
 271             totts.cwt = Rwave::cwt(ts, totnoct, nvoice, w0, plot=FALSE)
 272             ts.cwt = totts.cwt[,s0log:(s0log+noctave*nvoice)]
 273             #Normalization
 274             sqs <- sqrt(2^(0:(noctave*nvoice)/nvoice)*s0)
 275             sqres <- sweep(ts.cwt,2,sqs,'*')
 276             sqres / max(Mod(sqres))
 277         }
 278
 279         i = pair[1] ; j = pair[2]
 280         if (verbose && j==i+1)
 281             cat(paste("   Distances (",i,",",j,"), (",i,",",j+1,") ...\n", sep=""))
 282         cwt_i <- computeCWT(i)
 283         cwt_j <- computeCWT(j)
 284
 285 #print(system.time( {
 286         num <- epclustFilter(Mod(cwt_i * Conj(cwt_j)))
 287         WX  <- epclustFilter(Mod(cwt_i * Conj(cwt_i)))
 288         WY  <- epclustFilter(Mod(cwt_j * Conj(cwt_j)))
 289         wer2 <- sum(colSums(num)^2) / sum(colSums(WX) * colSums(WY))
 290         Xwer_dist[i,j] <- sqrt(delta * ncol(cwt_i) * max(1 - wer2, 0.)) #FIXME: wer2 should be < 1
 291         Xwer_dist[j,i] <- Xwer_dist[i,j]
 292 #} ) )
 293         Xwer_dist[i,i] = 0.
 294     }
 295
 296     if (parll)
 297     {
 298         cl = parallel::makeCluster(ncores_clust)
 299         synchrones_desc <- bigmemory::describe(synchrones)
 300         Xwer_dist_desc <- bigmemory::describe(Xwer_dist)
 301
 302         parallel::clusterExport(cl, varlist=c("synchrones_desc","Xwer_dist_desc","totnoct",
 303             "nvoice","w0","s0log","noctave","s0","verbose"), envir=environment())
 304     }
 305
 306     ignored <-
 307         if (parll)
 308             parallel::parLapply(cl, pairs, computeDistancesIJ)
 309         else
 310             lapply(pairs, computeDistancesIJ)
 311
 312     if (parll)
 313         parallel::stopCluster(cl)
 314
 315     Xwer_dist[n,n] = 0.
 316     distances <- Xwer_dist[,]
 317     rm(Xwer_dist) ; gc()
 318     distances #~small matrix K1 x K1
 319 }
 320
 321 # Helper function to divide indices into balanced sets
 322 .spreadIndices = function(indices, nb_per_chunk)
 323 {
 324     L = length(indices)
 325     nb_workers = floor( L / nb_per_chunk )
 326     if (nb_workers == 0)
 327     {
 328         # L < nb_series_per_chunk, simple case
 329         indices_workers = list(indices)
 330     }
 331     else
 332     {
 333         indices_workers = lapply( seq_len(nb_workers), function(i)
 334             indices[(nb_per_chunk*(i-1)+1):(nb_per_chunk*i)] )
 335         # Spread the remaining load among the workers
 336         rem = L %% nb_per_chunk
 337         while (rem > 0)
 338         {
 339             index = rem%%nb_workers + 1
 340             indices_workers[[index]] = c(indices_workers[[index]], indices[L-rem+1])
 341             rem = rem - 1
 342         }
 343     }
 344     indices_workers
 345 }