问题
I have the following code using ddply from plyr package:
ddply(mtcars,.(cyl),transform,freq=length(cyl))
The data.table version of this is :
DT<-data.table(mtcars)
DT[,freq:=.N,by=cyl]
How can I extend this when I have more than one function like the one below?
Now, I want to perform more than one function on ddply and data.table:
ddply(mtcars,.(cyl),transform,freq=length(cyl),sum=sum(mpg))
DT[,list(freq=.N,sum=sum(mpg)),by=cyl]
But, data.table gives me only three columns cyl,freq, and sum. Well, I can do like this:
DT[,list(freq=.N,sum=sum(mpg),mpg,disp,hp,drat,wt,qsec,vs,am,gear,carb),by=cyl]
But, I have large number of variables in my read data and I want all of them to be there as in ddply(...transform....). Is there shortcut in data.table just like doing := when we have only one function (as above) or something like this paste(names(mtcars),collapse=",") within data.table?
Note: I also have a large number of function to run. So, I can't repeat =: a number of times (but I would prefer this if lapply can be applied here).
回答1:
Use backquoted := like this...
DT[ , `:=`( freq = .N , sum = sum(mpg) ) , by=cyl ]
head( DT , 3 )
# mpg cyl disp hp drat wt qsec vs am gear carb freq sum
#1: 21.0 6 160 110 3.90 2.620 16.46 0 1 4 4 7 138.2
#2: 21.0 6 160 110 3.90 2.875 17.02 0 1 4 4 7 138.2
#3: 22.8 4 108 93 3.85 2.320 18.61 1 1 4 1 11 293.3
回答2:
Also useful in some situations:
newvars <- c("freq","sum")
DT[, `:=`(eval(newvars), list(.N,sum(mpg)))]
来源：https://stackoverflow.com/questions/19569145/equivalent-to-ddply-transform-in-data-table