draft final form of current package
[epclust.git] / epclust / R / main.R
CommitLineData
8702eb86 1#' CLAWS: CLustering with wAvelets and Wer distanceS
7f0781b7 2#'
56857861 3#' Groups electricity power curves (or any series of similar nature) by applying PAM
8702eb86
BA
4#' algorithm in parallel to chunks of size \code{nb_series_per_chunk}. Input series
5#' must be sampled on the same time grid, no missing values.
7f0781b7 6#'
8702eb86
BA
7#' @param getSeries Access to the (time-)series, which can be of one of the three
8#' following types:
9#' \itemize{
bf5c0844
BA
10#' \item [big.]matrix: each line contains all the values for one time-serie, ordered by time
11#' \item connection: any R connection object providing lines as described above
12#' \item character: name of a CSV file containing series in rows (no header)
8702eb86
BA
13#' \item function: a custom way to retrieve the curves; it has only one argument:
14#' the indices of the series to be retrieved. See examples
15#' }
4bcfdbee 16#' @inheritParams clustering
1c6f223e
BA
17#' @param K1 Number of super-consumers to be found after stage 1 (K1 << N)
18#' @param K2 Number of clusters to be found after stage 2 (K2 << K1)
4bcfdbee
BA
19#' @param wf Wavelet transform filter; see ?wavelets::wt.filter
20#' @param ctype Type of contribution: "relative" or "absolute" (or any prefix)
8702eb86
BA
21#' @param WER "end" to apply stage 2 after stage 1 has fully iterated, or "mix" to apply stage 2
22#' at the end of each task
4bcfdbee 23#' @param random TRUE (default) for random chunks repartition
1c6f223e
BA
24#' @param ntasks Number of tasks (parallel iterations to obtain K1 medoids); default: 1.
25#' Note: ntasks << N, so that N is "roughly divisible" by N (number of series)
5c652979
BA
26#' @param ncores_tasks "MPI" number of parallel tasks (1 to disable: sequential tasks)
27#' @param ncores_clust "OpenMP" number of parallel clusterings in one task
8702eb86
BA
28#' @param nb_series_per_chunk (~Maximum) number of series in each group, inside a task
29#' @param min_series_per_chunk Minimum number of series in each group
4bcfdbee 30#' @param sep Separator in CSV input file (if any provided)
8702eb86
BA
31#' @param nbytes Number of bytes to serialize a floating-point number; 4 or 8
32#' @param endian Endianness to use for (de)serialization. Use "little" or "big" for portability
4bcfdbee 33#' @param verbose Level of verbosity (0/FALSE for nothing or 1/TRUE for all; devel stage)
492cd9e7 34#' @param parll TRUE to fully parallelize; otherwise run sequentially (debug, comparison)
7f0781b7 35#'
bf5c0844 36#' @return A big.matrix of the final medoids curves (K2) in rows
1c6f223e
BA
37#'
38#' @examples
4efef8cc
BA
39#' \dontrun{
40#' # WER distances computations are a bit too long for CRAN (for now)
41#'
42#' # Random series around cos(x,2x,3x)/sin(x,2x,3x)
43#' x = seq(0,500,0.05)
44#' L = length(x) #10001
45#' ref_series = matrix( c(cos(x), cos(2*x), cos(3*x), sin(x), sin(2*x), sin(3*x)),
4bcfdbee 46#' byrow=TRUE, ncol=L )
4efef8cc
BA
47#' library(wmtsa)
48#' series = do.call( rbind, lapply( 1:6, function(i)
49#' do.call(rbind, wmtsa::wavBootstrap(ref_series[i,], n.realization=400)) ) )
50#' #dim(series) #c(2400,10001)
4bcfdbee 51#' medoids_ascii = claws(series, K1=60, K2=6, "d8", "rel", nb_series_per_chunk=500)
4efef8cc
BA
52#'
53#' # Same example, from CSV file
54#' csv_file = "/tmp/epclust_series.csv"
55#' write.table(series, csv_file, sep=",", row.names=FALSE, col.names=FALSE)
4bcfdbee 56#' medoids_csv = claws(csv_file, K1=60, K2=6, "d8", "rel", nb_series_per_chunk=500)
4efef8cc
BA
57#'
58#' # Same example, from binary file
59#' bin_file = "/tmp/epclust_series.bin"
60#' nbytes = 8
61#' endian = "little"
4bcfdbee 62#' epclust::binarize(csv_file, bin_file, 500, nbytes, endian)
4efef8cc 63#' getSeries = function(indices) getDataInFile(indices, bin_file, nbytes, endian)
4bcfdbee 64#' medoids_bin = claws(getSeries, K1=60, K2=6, "d8", "rel", nb_series_per_chunk=500)
4efef8cc
BA
65#' unlink(csv_file)
66#' unlink(bin_file)
67#'
68#' # Same example, from SQLite database
69#' library(DBI)
70#' series_db <- dbConnect(RSQLite::SQLite(), "file::memory:")
71#' # Prepare data.frame in DB-format
72#' n = nrow(series)
4bcfdbee
BA
73#' time_values = data.frame(
74#' id = rep(1:n,each=L),
75#' time = rep( as.POSIXct(1800*(0:n),"GMT",origin="2001-01-01"), L ),
76#' value = as.double(t(series)) )
4efef8cc 77#' dbWriteTable(series_db, "times_values", times_values)
4bcfdbee
BA
78#' # Fill associative array, map index to identifier
79#' indexToID_inDB <- as.character(
80#' dbGetQuery(series_db, 'SELECT DISTINCT id FROM time_values')[,"id"] )
4efef8cc 81#' getSeries = function(indices) {
4bcfdbee
BA
82#' request = "SELECT id,value FROM times_values WHERE id in ("
83#' for (i in indices)
84#' request = paste(request, i, ",", sep="")
85#' request = paste(request, ")", sep="")
86#' df_series = dbGetQuery(series_db, request)
87#' # Assume that all series share same length at this stage
88#' ts_length = sum(df_series[,"id"] == df_series[1,"id"])
89#' t( as.matrix(df_series[,"value"], nrow=ts_length) )
4efef8cc 90#' }
4bcfdbee
BA
91#' medoids_db = claws(getSeries, K1=60, K2=6, "d8", "rel", nb_series_per_chunk=500)
92#' dbDisconnect(series_db)
93#'
94#' # All computed medoids should be the same:
95#' digest::sha1(medoids_ascii)
96#' digest::sha1(medoids_csv)
97#' digest::sha1(medoids_bin)
98#' digest::sha1(medoids_db)
1c6f223e 99#' }
1c6f223e 100#' @export
56857861 101claws = function(getSeries, K1, K2,
4bcfdbee 102 wf,ctype, #stage 1
56857861 103 WER="end", #stage 2
4bcfdbee 104 random=TRUE, #randomize series order?
56857861
BA
105 ntasks=1, ncores_tasks=1, ncores_clust=4, #control parallelism
106 nb_series_per_chunk=50*K1, min_series_per_chunk=5*K1, #chunk size
107 sep=",", #ASCII input separator
4bcfdbee 108 nbytes=4, endian=.Platform$endian, #serialization (write,read)
492cd9e7 109 verbose=FALSE, parll=TRUE)
ac1d4231 110{
0e2dce80 111 # Check/transform arguments
492cd9e7
BA
112 if (!is.matrix(getSeries) && !bigmemory::is.big.matrix(getSeries)
113 && !is.function(getSeries)
114 && !methods::is(getSeries,"connection") && !is.character(getSeries))
0e2dce80 115 {
492cd9e7 116 stop("'getSeries': [big]matrix, function, file or valid connection (no NA)")
5c652979 117 }
56857861
BA
118 K1 = .toInteger(K1, function(x) x>=2)
119 K2 = .toInteger(K2, function(x) x>=2)
120 if (!is.logical(random))
121 stop("'random': logical")
122 tryCatch(
4bcfdbee 123 {ignored <- wavelets::wt.filter(wf)},
56857861 124 error = function(e) stop("Invalid wavelet filter; see ?wavelets::wt.filter"))
7f0781b7
BA
125 if (WER!="end" && WER!="mix")
126 stop("WER takes values in {'end','mix'}")
56857861
BA
127 ntasks = .toInteger(ntasks, function(x) x>=1)
128 ncores_tasks = .toInteger(ncores_tasks, function(x) x>=1)
129 ncores_clust = .toInteger(ncores_clust, function(x) x>=1)
130 nb_series_per_chunk = .toInteger(nb_series_per_chunk, function(x) x>=K1)
131 min_series_per_chunk = .toInteger(K1, function(x) x>=K1 && x<=nb_series_per_chunk)
132 if (!is.character(sep))
133 stop("'sep': character")
134 nbytes = .toInteger(nbytes, function(x) x==4 || x==8)
135
136 # Serialize series if required, to always use a function
dc646a97 137 bin_dir = ".epclust_bin/"
56857861
BA
138 dir.create(bin_dir, showWarnings=FALSE, mode="0755")
139 if (!is.function(getSeries))
140 {
4bcfdbee
BA
141 if (verbose)
142 cat("...Serialize time-series\n")
56857861 143 series_file = paste(bin_dir,"data",sep="") ; unlink(series_file)
4bcfdbee
BA
144 binarize(getSeries, series_file, nb_series_per_chunk, sep, nbytes, endian)
145 getSeries = function(inds) getDataInFile(inds, series_file, nbytes, endian)
56857861 146 }
ac1d4231 147
95b5c2e6 148 # Serialize all computed wavelets contributions into a file
4bcfdbee 149 contribs_file = paste(bin_dir,"contribs",sep="") ; unlink(contribs_file)
7f0781b7 150 index = 1
cea14f3a 151 nb_curves = 0
4bcfdbee
BA
152 if (verbose)
153 cat("...Compute contributions and serialize them\n")
492cd9e7
BA
154 nb_curves = binarizeTransform(getSeries,
155 function(series) curvesToContribs(series, wf, ctype),
156 contribs_file, nb_series_per_chunk, nbytes, endian)
4bcfdbee 157 getContribs = function(indices) getDataInFile(indices, contribs_file, nbytes, endian)
8e6accca 158
5c652979
BA
159 if (nb_curves < min_series_per_chunk)
160 stop("Not enough data: less rows than min_series_per_chunk!")
161 nb_series_per_task = round(nb_curves / ntasks)
162 if (nb_series_per_task < min_series_per_chunk)
163 stop("Too many tasks: less series in one task than min_series_per_chunk!")
ac1d4231 164
492cd9e7
BA
165 runTwoStepClustering = function(inds)
166 {
bf5c0844 167 if (parll && ntasks>1)
492cd9e7
BA
168 require("epclust", quietly=TRUE)
169 indices_medoids = clusteringTask1(
170 inds, getContribs, K1, nb_series_per_chunk, ncores_clust, verbose, parll)
56857861
BA
171 if (WER=="mix")
172 {
363ae134 173 require("bigmemory", quietly=TRUE)
bf5c0844 174 medoids1 = bigmemory::as.big.matrix( getSeries(indices_medoids) )
a174b8ea
BA
175 medoids2 = clusteringTask2(medoids1, K2, getSeries, nb_curves, nb_series_per_chunk,
176 nbytes, endian, ncores_clust, verbose, parll)
4bcfdbee 177 binarize(medoids2, synchrones_file, nb_series_per_chunk, sep, nbytes, endian)
56857861
BA
178 return (vector("integer",0))
179 }
180 indices_medoids
492cd9e7
BA
181 }
182
c45fd663
BA
183 # Cluster contributions in parallel (by nb_series_per_chunk)
184 indices_all = if (random) sample(nb_curves) else seq_len(nb_curves)
185 indices_tasks = lapply(seq_len(ntasks), function(i) {
186 upper_bound = ifelse( i<ntasks, min(nb_series_per_task*i,nb_curves), nb_curves )
187 indices_all[((i-1)*nb_series_per_task+1):upper_bound]
188 })
189 if (verbose)
e161499b
BA
190 {
191 message = paste("...Run ",ntasks," x stage 1", sep="")
192 if (WER=="mix")
193 message = paste(message," + stage 2", sep="")
194 cat(paste(message,"\n", sep=""))
195 }
c45fd663
BA
196 if (WER=="mix")
197 {synchrones_file = paste(bin_dir,"synchrones",sep="") ; unlink(synchrones_file)}
bf5c0844 198 if (parll && ntasks>1)
c45fd663
BA
199 {
200 cl = parallel::makeCluster(ncores_tasks)
201 varlist = c("getSeries","getContribs","K1","K2","verbose","parll",
bf5c0844 202 "nb_series_per_chunk","ntasks","ncores_clust","sep","nbytes","endian")
c45fd663
BA
203 if (WER=="mix")
204 varlist = c(varlist, "synchrones_file")
205 parallel::clusterExport(cl, varlist=varlist, envir = environment())
206 }
207
492cd9e7 208 # 1000*K1 indices [if WER=="end"], or empty vector [if WER=="mix"] --> series on file
bf5c0844 209 if (parll && ntasks>1)
492cd9e7
BA
210 indices = unlist( parallel::parLapply(cl, indices_tasks, runTwoStepClustering) )
211 else
212 indices = unlist( lapply(indices_tasks, runTwoStepClustering) )
bf5c0844 213 if (parll && ntasks>1)
492cd9e7 214 parallel::stopCluster(cl)
3465b246 215
8702eb86 216 getRefSeries = getSeries
e205f218
BA
217 if (WER=="mix")
218 {
219 indices = seq_len(ntasks*K2)
220 #Now series must be retrieved from synchrones_file
56857861 221 getSeries = function(inds) getDataInFile(inds, synchrones_file, nbytes, endian)
4bcfdbee
BA
222 #Contributions must be re-computed
223 unlink(contribs_file)
e205f218 224 index = 1
4bcfdbee
BA
225 if (verbose)
226 cat("...Serialize contributions computed on synchrones\n")
492cd9e7
BA
227 ignored = binarizeTransform(getSeries,
228 function(series) curvesToContribs(series, wf, ctype),
229 contribs_file, nb_series_per_chunk, nbytes, endian)
e205f218 230 }
0e2dce80
BA
231
232 # Run step2 on resulting indices or series (from file)
4bcfdbee
BA
233 if (verbose)
234 cat("...Run final // stage 1 + stage 2\n")
492cd9e7 235 indices_medoids = clusteringTask1(
af3ea947 236 indices, getContribs, K1, nb_series_per_chunk, ncores_tasks*ncores_clust, verbose, parll)
bf5c0844 237 medoids1 = bigmemory::as.big.matrix( getSeries(indices_medoids) )
a174b8ea
BA
238 medoids2 = clusteringTask2(medoids1, K2, getRefSeries, nb_curves, nb_series_per_chunk,
239 nbytes, endian, ncores_tasks*ncores_clust, verbose, parll)
4bcfdbee
BA
240
241 # Cleanup
242 unlink(bin_dir, recursive=TRUE)
243
bf5c0844 244 medoids2
56857861
BA
245}
246
4bcfdbee
BA
247#' curvesToContribs
248#'
249#' Compute the discrete wavelet coefficients for each series, and aggregate them in
250#' energy contribution across scales as described in https://arxiv.org/abs/1101.4744v2
251#'
252#' @param series Matrix of series (in rows), of size n x L
253#' @inheritParams claws
254#'
255#' @return A matrix of size n x log(L) containing contributions in rows
256#'
257#' @export
258curvesToContribs = function(series, wf, ctype)
56857861
BA
259{
260 L = length(series[1,])
261 D = ceiling( log2(L) )
262 nb_sample_points = 2^D
4bcfdbee
BA
263 cont_types = c("relative","absolute")
264 ctype = cont_types[ pmatch(ctype,cont_types) ]
8702eb86 265 t( apply(series, 1, function(x) {
56857861
BA
266 interpolated_curve = spline(1:L, x, n=nb_sample_points)$y
267 W = wavelets::dwt(interpolated_curve, filter=wf, D)@W
4bcfdbee
BA
268 nrj = rev( sapply( W, function(v) ( sqrt( sum(v^2) ) ) ) )
269 if (ctype=="relative") nrj / sum(nrj) else nrj
8702eb86 270 }) )
56857861
BA
271}
272
492cd9e7 273# Check integer arguments with functional conditions
56857861
BA
274.toInteger <- function(x, condition)
275{
276 if (!is.integer(x))
277 tryCatch(
278 {x = as.integer(x)[1]},
279 error = function(e) paste("Cannot convert argument",substitute(x),"to integer")
280 )
281 if (!condition(x))
282 stop(paste("Argument",substitute(x),"does not verify condition",body(condition)))
283 x
cea14f3a 284}