isSubset.R
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
library(foreach) | |
library(doParallel) | |
library(arules) | |
library(Rcpp) | |
library(isSub) #this is our Rcpp library | |
#register cores | |
registerDoParallel(3) | |
#function searching for a subsets | |
issub=function(data){ | |
issub=rep(0,nrow(data)) | |
Antecedent=data$lhs | |
Consequent=data$rhs | |
length=nrow(data) | |
for(i in 1:length){ | |
for(j in (i+1):length){ | |
if(isTRUE(grepl(Antecedent[j], Antecedent[i]))){ | |
issub[i]=1 | |
} | |
else if(isTRUE(grepl(Antecedent[i], Antecedent[j]))){ | |
issub[j]=1 | |
} | |
} | |
} | |
data$isSubset=issub | |
return(data) | |
} | |
#change datatype of rhs and lhs to character, removing '{}' | |
data$lhs=as.character(data$lhs) | |
data$rhs=as.character(data$rhs) | |
data$lhs=gsub('[{}]',replace='',data$lhs) | |
data$rhs=gsub('[{}]',replace='',data$rhs) | |
#create table of unique consequents-used for parallel computation with foreach | |
listOfUniqueConsequents=table(data$rhs) | |
#here calculation are performed sequentially, after division of the dataset into subgroups of unique consequents | |
resultsSequentiallyR=c() | |
resultsSequentiallyRcpp=c() | |
for(i in 1:length(listOfUniqueConsequents)){ | |
subset=data[which(data$rhs==as.character(as.integer(names(listOfUniqueConsequents[i])))),] | |
resultsSequentiallyR=c(resultsSequentiallyR,issub(subset)) | |
resultsSequentiallyRcpp=c(resultsSequentiallyRcpp,isSubset(subset)) | |
} | |
#foreach with Rcpp function-remember to include package, otherwise function would not be exported | |
resultsParallelRcpp=foreach(k=1:length(listOfUniqueConsequents),.packages=c("isSub"))%dopar%{isSubset(data[which(data$rhs==as.character(as.integer(names(listOfUniqueConsequents[k])))),])} | |
#foreach with R function | |
resultsParallelR=foreach(k=1:length(listOfUniqueConsequents))%dopar%{issub(data[which(data$rhs==as.character(as.integer(names(listOfUniqueConsequents[k])))),])} | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment