SparkR create table RelativeFrequency - r

Question

Hi I'm working on SparkR. I'm try to calculate a RelativeFrequency of my Data.
SmsInt<-fread("smsCallInt.txt")
setnames(SmsInt,c("V1","V2","V3","V4","V5","V6","V7","V8"),
c("SquareID","TimeInterval","CountryCode","SmsIn","SmsOut","CallIn","CallOut","Internet"))
#Also create a dataFrame from it.
SmsInt$TimeInterval<-as.numeric(SmsInt$TimeInterval)
SmsInt.df<-createDataFrame(sqlContext,SmsInt[1:500,])
str(SmsInt)
Classes ‘data.table’ and 'data.frame': 2459324 obs. of 8 variables:
$ SquareID : int 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 ...
$ TimeInterval: num 1.38e+12 1.38e+12 1.38e+12 1.38e+12 1.38e+12 ...
$ CountryCode : int 0 39 49 0 39 0 39 0 39 49 ...
$ SmsIn : num 0.109 1.001 NA 0.193 0.648 ...
$ SmsOut : num NA 1.26 NA NA 1.06 ...
$ CallIn : num NA 0.0876 NA NA 0.1751 ...
$ CallOut : num 0.0219 0.2196 NA NA 0.1532 ...
$ Internet : num NA 10.1685 0.0219 NA 11.8671 ...
- attr(*, ".internal.selfref")=<externalptr>
What I want to do is create a RelativeFrequency from SmsInt$CountryCode.
When I type Country<-table(SmsInt$CountryCode)
I got this Error:
Errore: class(objId) == "jobj" is not TRUE
What can I do?There is a way to calculate it manually or with some package?
I created an algorithm but i have some trouble .
Country5<-SmsInt$CountryCode[1:90]
UniqueCountry<-unique(Country5)
VectorLen<-c()
Parsed<-c()
Freq<-c()
for(i in 1:length(UniqueCountry)){
CountryCode.i<-UniqueCountry[i]
if(CountryCode.i %in% Parsed){
Vector<-0
VectorLen[i]<-0
Freq[i]<-0
}
else{
Vector<-grep(CountryCode.i,Country5)
Parsed[i]<-CountryCode.i
VectorLen[i]<-length(Vector)
Freq[i]<-VectorLen[i]/90
Vector<-0
}
}
Vector
VectorLen #92 it needs to be 90
Freq
sum(Freq) #1.022222 needs to be 1
With 80 all works.

Answer 1

Ok I did it. The error was the grep function so when I was looking for number 1 it was found on number 10 once, for example.
I post the solution here.
RelativeFrequency<-function(DataSet){
UniqueCountry<-unique(DataSet)
VectorLen<-c()
Parsed<-c()
Freq<-c()
for(i in 1:length(UniqueCountry)){
CountryCode.i<-UniqueCountry[i]
if(CountryCode.i %in% Parsed){
Vector<-0
VectorLen[i]<-0
Freq[i]<-0
}
else{
Vector<-which(DataSet %in% CountryCode.i)
Parsed[i]<-CountryCode.i
VectorLen[i]<-length(Vector)
Freq[i]<-VectorLen[i]/length(DataSet)
}
}
print("Vector of RelativeFrequency")
print(Freq)
print("Frequency Sum (Needs to be 1)")
print(sum(Freq))
print("Parsed element ")
print(Parsed)
barplot(Freq,names=Parsed,space = 0.7,axisnames = TRUE,las=2)
}