From: Benjamin Auder Date: Thu, 9 Mar 2017 10:58:25 +0000 (+0100) Subject: fix typo, add some TODO X-Git-Url: https://git.auder.net/variants/current/doc/scripts/%7B%7B%20pkg.url%20%7D%7D?a=commitdiff_plain;h=2c14dbea13c897c6964f49f9cd17622f4c9733c0;p=epclust.git fix typo, add some TODO --- diff --git a/.gitignore b/.gitignore index c946480..c8ea425 100644 --- a/.gitignore +++ b/.gitignore @@ -40,3 +40,7 @@ #ignore RcppExports, generated by Rcpp::compileAttributes /epclust/R/RcppExports.R /epclust/src/RcppExports.cpp + +#misc +Rprof.out +*.zip diff --git a/TODO b/TODO index 199a59f..4f865b0 100644 --- a/TODO +++ b/TODO @@ -47,3 +47,15 @@ utiliser Rcpp ? #' @importFrom synchronicity boost.mutex lock unlock subtree: epclust, shared. This root folder should remain private + +#TODO: use dbs(), + #https://www.r-bloggers.com/debugging-parallel-code-with-dbs/ + #http://gforge.se/2015/02/how-to-go-parallel-in-r-basics-tips/ + +synchrones --> somme, pas moyenne + +PLOT: +plot manifold 2D distances WER / +fenetre tempo forme des courbes / +medoids / +gain en prevision: clust puis full --> enercast diff --git a/biblio/Clustering_Functional_Data_Using_Wavelets-Antoniadis2013.pdf b/biblio/Clustering_Functional_Data_Using_Wavelets-Antoniadis2013.pdf new file mode 100644 index 0000000..5801a64 --- /dev/null +++ b/biblio/Clustering_Functional_Data_Using_Wavelets-Antoniadis2013.pdf @@ -0,0 +1 @@ +#$# git-fat 7571c6c3b737bf2f57eab62fea99eb24a756eded 895937 diff --git a/epclust/R/clustering.R b/epclust/R/clustering.R index 4d43b2b..7e06c43 100644 --- a/epclust/R/clustering.R +++ b/epclust/R/clustering.R @@ -134,29 +134,8 @@ computeSynchrones = function(medoids, getRefSeries, m <- synchronicity::attach.mutex(m_desc) } - - -#TODO: use dbs(), - #https://www.r-bloggers.com/debugging-parallel-code-with-dbs/ - #http://gforge.se/2015/02/how-to-go-parallel-in-r-basics-tips/ - -#OK :: -#write(length(indices), file="TOTO") -#write( computeMedoidsIndices(medoids@address, getRefSeries(indices[1:600])), file="TOTO") -#stop() - -# write(indices, file="TOTO", ncolumns=10, append=TRUE) -#write("medoids", file = "TOTO", ncolumns=1, append=TRUE) -#write(medoids[1,1:3], file = "TOTO", ncolumns=1, append=TRUE) -#write("synchrones", file = "TOTO", ncolumns=1, append=TRUE) -#write(synchrones[1,1:3], file = "TOTO", ncolumns=1, append=TRUE) - -#NOT OK :: (should just be "ref_series") ...or yes ? race problems mutex then ? ?! #get medoids indices for this chunk of series - mi = computeMedoidsIndices(medoids@address, getRefSeries(indices[1:600])) #ref_series) -write("MI ::::", file = "TOTO", ncolumns=1, append=TRUE) -write(mi[1:3], file = "TOTO", ncolumns=1, append=TRUE) - + mi = computeMedoidsIndices(medoids@address, ref_series) # #R-equivalent, requiring a matrix (thus potentially breaking "fit-in-memory" hope) # mat_meds = medoids[,] # mi = rep(NA,nb_series) @@ -169,6 +148,7 @@ write(mi[1:3], file = "TOTO", ncolumns=1, append=TRUE) if (parll) synchronicity::lock(m) synchrones[mi[i],] = synchrones[mi[i],] + ref_series[i,] +#TODO: remove counts counts[mi[i],1] = counts[mi[i],1] + 1 if (parll) synchronicity::unlock(m) @@ -266,14 +246,17 @@ computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE) # Distance between rows i and j computeDistancesIJ = function(pair) { - require("bigmemory", quietly=TRUE) - require("epclust", quietly=TRUE) - synchrones <- bigmemory::attach.big.matrix(synchrones_desc) - Xwer_dist <- bigmemory::attach.big.matrix(Xwer_dist_desc) - - computeCWT = function(i) + if (parll) { - ts <- scale(ts(synchrones[i,]), center=TRUE, scale=scaled) + require("bigmemory", quietly=TRUE) + require("epclust", quietly=TRUE) + synchrones <- bigmemory::attach.big.matrix(synchrones_desc) + Xwer_dist <- bigmemory::attach.big.matrix(Xwer_dist_desc) + } + + computeCWT = function(index) + { + ts <- scale(ts(synchrones[index,]), center=TRUE, scale=scaled) totts.cwt = Rwave::cwt(ts, totnoct, nvoice, w0, plot=FALSE) ts.cwt = totts.cwt[,s0log:(s0log+noctave*nvoice)] #Normalization @@ -281,18 +264,21 @@ computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE) sqres <- sweep(ts.cwt,2,sqs,'*') sqres / max(Mod(sqres)) } - +#browser() i = pair[1] ; j = pair[2] if (verbose && j==i+1) cat(paste(" Distances (",i,",",j,"), (",i,",",j+1,") ...\n", sep="")) - cwt_i = computeCWT(i) - cwt_j = computeCWT(j) +print(system.time( { cwt_i <- computeCWT(i) + cwt_j <- computeCWT(j) } )) + +print(system.time( { num <- epclustFilter(Mod(cwt_i * Conj(cwt_j))) 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) * (1 - wer2)) + Xwer_dist[i,j] <- sqrt(delta * ncol(cwt_i) * max(1 - wer2, 0.)) #FIXME: wer2 should be < 1 Xwer_dist[j,i] <- Xwer_dist[i,j] +} ) ) Xwer_dist[i,i] = 0. } @@ -300,12 +286,12 @@ computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE) { cl = parallel::makeCluster(ncores_clust) synchrones_desc <- bigmemory::describe(synchrones) - Xwer_dist_desc_desc <- bigmemory::describe(Xwer_dist) + Xwer_dist_desc <- bigmemory::describe(Xwer_dist) parallel::clusterExport(cl, varlist=c("synchrones_desc","Xwer_dist_desc","totnoct", "nvoice","w0","s0log","noctave","s0","verbose"), envir=environment()) } - +browser() ignored <- if (parll) parallel::parLapply(cl, pairs, computeDistancesIJ) @@ -314,7 +300,7 @@ computeWerDists = function(synchrones, ncores_clust=1,verbose=FALSE,parll=TRUE) if (parll) parallel::stopCluster(cl) - +#browser() Xwer_dist[n,n] = 0. distances <- Xwer_dist[,] rm(Xwer_dist) ; gc()