separate TODO and wrapper example

[epclust.git] / old_C_code / wrapper.R

#Exemple :
#
#dans old_C_code/build :
#cmake ../stage1/src
#make
#
#dans data/, lancer R puis :
#source("../old_C_code/wrapper.R")
#serialize("../old_C_code/build", "2009.csv","2009.bin",1)
#library(parallel)
#np = detectCores()
#nbSeriesPerChunk = 3000
#nbClusters = 20
#ppam_exe("../old_C_code/build",np,"2009.bin",nbSeriesPerChunk,nbClusters)
#C = getMedoids("../old_C_code/build", "ppamResult.xml", "ppamFinalSeries.bin")
#first100series = deserialize("../old_C_code/build", "2009.bin", "2009.csv.part", "1-100")
#distor = getDistor("../old_C_code/build", "ppamResult.xml", "2009.bin")

ppam_exe = function(path=".", np=parallel::detectCores(), data=NULL,
	nbSeriesPerChunk, nbClusters, randomize=1, p_dissims=2)
{
	args = paste(nbSeriesPerChunk," ",nbClusters," ",randomize," ",p_dissims,sep="")

	command_line = paste("mpirun -np ",np," ",path,"/ppam.exe cluster",sep="")

	#if data provided (as data.frame or matrix...): binarize it, and add it as first argument
	if (!is.null(data))
	{
		if (!is.character(data))
		{
			#assuming matrix or data.frame, WITH row names
			#( identifiers; could be line number... e.g. data <- cbind(1:nrow(data),data) )
			write.table(data, "/tmp/data_csv", sep=",", row.names=FALSE, col.names=FALSE)
			system(paste(path,"/ppam.exe serialize /tmp/data_csv /tmp/data_bin 0 0",sep=""))
		} else
		{
			system(paste(path,"/ppam.exe serialize ",data," /tmp/data_bin 0 0",sep=""))
		}
		command_line = paste(command_line," /tmp/data_bin",sep="")
	}

	command_line = paste(command_line," ",args,sep="")
	system(command_line)
}

#NOTE: identifiers in first column
getMedoids = function(path=".", xmlResult = "ppamResult.xml",
	finalSeries = "ppamFinalSeries.bin")
{
	system(paste(path,"/ppam.exe deserialize ",finalSeries," ppamFinalSeries.csv -1",sep=""))
	curves = read.table("ppamFinalSeries.csv", sep=",")
	library(XML)
	ranks = as.integer( xmlToList( xmlParse(xmlResult) )$ranks )
	return ( curves[ranks,] ) # == medoids
}

getDistor = function(path=".", xmlResult = "ppamResult.xml",
	finalSeries = "ppamFinalSeries.bin")
{
	system(paste(path,"/ppam.exe classif ",finalSeries," ",xmlResult,sep=""))
}

serialize = function(path=".", csvSeries, binSeries, byCols=0, nbSeries=0)
{
	system(paste(path,"/ppam.exe serialize ",csvSeries," ",binSeries," ",byCols," ",nbSeries,
		sep=""))
}

deserialize = function(path=".", binSeries, csvSeries, ranks="-1", return=TRUE)
{
	system(paste(path,"/ppam.exe deserialize ",binSeries," ",csvSeries," ",ranks,sep=""))
	if (return)
		return ( read.table(csvSeries, sep=",") )
}