CODEWITHSUNDEEP
rapplylapply
sjgknight
sjgknight

Reputation: 415

R store output from lapply with multiple functions

I'm using lapply to loop through a list of dataframes and apply the same set of functions. This works fine when lapply has just one function, but I'm struggling to see how I store/print the output from multiple functions - in that case, I seem to only get output from one 'loop'.

So this:

output <- lapply(dflis,function(lismember) vss(ISEQData,n=9,rotate="oblimin",diagonal=F,fm="ml"))

works, while the following doesn't:

output <- lapply(dflis,function(lismember){
outputvss <- vss(lismember,n=9,rotate="oblimin",diagonal=F,fm="ml")
nefa <- (EFA.Comp.Data(Data=lismember, F.Max=9, Graph=T))
})

I think this dummy example is an analogue, so in other words:

nbs <- list(1==1,2==2,3==3,4==4)
nbsout <- lapply(nbs,function(x) length(x))

Gives me something I can access, while I can't see how to store output using the below (e.g. the attempt to use nbsout[[x]][2]):

nbs <- list(1==1,2==2,3==3,4==4)
nbsout <- lapply(nbs,function(x){
  nbsout[[x]][1]<-typeof(x)
  nbsout[[x]][2]<-length(x)
  }
                 )

I'm using RStudio and will then be printing outputs/knitting html (where it makes sense to display the results from each dataset together, rather than each function-output for each dataset sequentially).

Upvotes: 2

Views: 5466

Answers (2)

agstudy
agstudy

Reputation: 121568

You should return a structure that include all your outputs. Better to return a named list. You can also return a data.frame if your outputs have all the same dimensions.

otutput <- lapply(dflis,function(lismember){
    outputvss <- vss(lismember,n=9,rotate="oblimin",diagonal=F,fm="ml")
    nefa <- (EFA.Comp.Data(Data=lismember, F.Max=9, Graph=T))

    list(outputvss=outputvss,nefa=nefa)  
    ## or data.frame(outputvss=outputvss,nefa=nefa) 
})

When you return a data.frame you can use sapply that simply outputs the final result to a big data.frame. Or you can use the classical: 

do.call(rbind,output)

to aggregate the result.

Upvotes: 3

Roland
Roland

Reputation: 132606

A function should always have an explicit return value, e.g.

output <- lapply(dflis,function(lismember){
   outputvss <- vss(lismember,n=9,rotate="oblimin",diagonal=F,fm="ml")
   nefa <- (EFA.Comp.Data(Data=lismember, F.Max=9, Graph=T))
   #return value:
   list(outputvss, nefa)
})

output is then a list of lists.

Upvotes: 2

Related Questions