export vars to nodes

[epclust.git] / epclust / R / main.R

diff --git a/epclust/R/main.R b/epclust/R/main.R
index f5ad81a..75041a4 100644 (file)
--- a/epclust/R/main.R
+++ b/epclust/R/main.R
@@ -67,8 +67,7 @@ epclust = function(data, K1, K2, ntasks=1, nb_series_per_chunk=50*K1, min_series
                stop("WER takes values in {'end','mix'}")
 
        # Serialize all wavelets coefficients (+ IDs) onto a file
-       coeffs_file = ".coeffs"
-       ids_files = ".ids"
+       unlink(".coeffs")
        index = 1
        nb_curves = 0
        nb_coeffs = NA
@@ -77,7 +76,7 @@ epclust = function(data, K1, K2, ntasks=1, nb_series_per_chunk=50*K1, min_series
                coeffs_chunk = computeCoeffs(data, index, nb_series_per_chunk, wf)
                if (is.null(coeffs_chunk))
                        break
-               serialize(coeffs_chunk, coeffs_file, append=TRUE)
+               writeCoeffs(coeffs_chunk)
                index = index + nb_series_per_chunk
                nb_curves = nb_curves + nrow(coeffs_chunk)
                if (is.na(nb_coeffs))
@@ -91,22 +90,21 @@ epclust = function(data, K1, K2, ntasks=1, nb_series_per_chunk=50*K1, min_series
                stop("Too many tasks: less series in one task than min_series_per_chunk!")
 
        # Cluster coefficients in parallel (by nb_series_per_chunk)
-       indices = if (random) sample(nb_curves) else seq_len(nb_curves) #all indices
-       indices_tasks = list() #indices to be processed in each task
-       for (i in seq_len(ntasks))
-       {
+       indices = if (random) sample(nb_curves) else seq_len(nb_curves)
+       indices_tasks = lapply(seq_len(ntasks), function(i) {
                upper_bound = ifelse( i<ntasks, min(nb_series_per_task*i,nb_curves), nb_curves )
-               indices_task[[i]] = indices[((i-1)*nb_series_per_task+1):upper_bound]
-       }
+               indices[((i-1)*nb_series_per_task+1):upper_bound]
+       })
        library(parallel, quietly=TRUE)
        cl_tasks = parallel::makeCluster(ncores_tasks)
-       #parallel::clusterExport(cl=cl_tasks, varlist=c("ncores_clust", ...), envir=environment())
-       indices = parallel::parLapply(cl_tasks, indices_tasks, clusteringStep12, )
+       parallel::clusterExport(cl_tasks,
+               varlist=c("K1","K2","WER","nb_series_per_chunk","ncores_clust"),#TODO: pass also
+                                               #nb_coeffs...and filename (in a list... ?)
+               envir=environment())
+       indices = parallel::parLapply(cl_tasks, indices_tasks, clusteringTask)
        parallel::stopCluster(cl_tasks)
 
-##TODO: passer data ?!
-
        # Run step1+2 step on resulting ranks
-       ranks = clusteringStep12()
-       return (list("ranks"=ranks, "medoids"=getSeries(data, ranks)))
+       indices = clusterChunk(indices, K1, K2)
+       return (list("indices"=indices, "medoids"=getSeries(data, indices)))
 }