de_serialize works. Variables names improved. Code beautified. TODO: clustering tests

[epclust.git] / epclust / R / clustering.R

diff --git a/epclust/R/clustering.R b/epclust/R/clustering.R
index c8bad66..fce1b1c 100644 (file)
--- a/epclust/R/clustering.R
+++ b/epclust/R/clustering.R
@@ -1,6 +1,5 @@
-# Cluster one full task (nb_curves / ntasks series)
-clusteringTask = function(indices,getSeries,getSeriesForSynchrones,synchrones_file,
-       getCoefs,K1,K2,nb_series_per_chunk,ncores,to_file)
+# 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
@@ -12,46 +11,51 @@ clusteringTask = function(indices,getSeries,getSeriesForSynchrones,synchrones_fi
                        indices[(nb_series_per_chunk*(i-1)+1):upper_bound]
                })
                indices = unlist( parallel::parLapply(cl, indices_workers, function(inds)
-                       computeClusters1(inds, getCoefs, K1)) )
-               if (length(indices_clust) == K1)
+                       computeClusters1(getCoefs(inds), K1)) )
+               if (length(indices) == K1)
                        break
        }
        parallel::stopCluster(cl)
-       if (K2 == 0)
-               return (indices)
-       computeClusters2(indices, K2, getSeries, getSeriesForSynchrones, to_file)
-       vector("integer",0)
+       indices #medoids
 }
 
 # Apply the clustering algorithm (PAM) on a coeffs or distances matrix
-computeClusters1 = function(indices, getCoefs, K1)
-{
-       coefs = getCoefs(indices)
+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(indices, K2, getSeries, getSeriesForSynchrones, to_file)
+computeClusters2 = function(medoids, K2, getRefSeries, nb_series_per_chunk)
 {
-       curves = computeSynchrones(indices, getSeries, getSeriesForSynchrones)
-       dists = computeWerDists(curves)
-       medoids = cluster::pam(dists, K2, diss=TRUE)$medoids
-       if (to_file)
-       {
-               serialize(medoids, synchrones_file)
-               return (NULL)
-       }
-       medoids
+       synchrones = computeSynchrones(medoids, getRefSeries, nb_series_per_chunk)
+       cluster::pam(computeWerDists(synchrones), K2, diss=TRUE)$medoids
 }
 
 # Compute the synchrones curves (sum of clusters elements) from a clustering result
-computeSynchrones = function(indices, getSeries, getSeriesForSynchrones)
+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 ?
-       series = getSeries(indices)
-       #...........
-       #sapply(seq_along(inds), colMeans(getSeries(inds[[i]]$indices,inds[[i]]$ids)))
+       K = nrow(medoids)
+       synchrones = matrix(0, nrow=K, ncol=ncol(medoids))
+       counts = rep(0,K)
+       index = 1
+       repeat
+       {
+               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)))
+               {
+                       j = which.min( rowSums( sweep(medoids, 2, series[i,], '-')^2 ) )
+                       synchrones[j,] = synchrones[j,] + series[i,]
+                       counts[j] = counts[j] + 1
+               }
+               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, '/')
 }
 
 # Compute the WER distance between the synchrones curves (in rows)