Skip to content

How to duplicate each row in a data.frame?

4 messages · Peng Yu, Phil Spector, David Freedman +1 more

Original
Peng Yu
# 
I want to duplicate each line in 'df' 3 times. But I'm confused why
'z' is a 6 by 4 matrix. Could somebody let me know what the correct
way is to duplicate each row of a data.frame?

df=expand.grid(x1=c('a','b'),x2=c('u','v'))
n=3
z=apply(df,1
    ,function(x){
      result=do.call(rbind,rep(list(x),n))
      result
    }
    )
z
Phil Spector
# 
df[rep(1:nrow(df),each=3),]

 					- Phil Spector
 					 Statistical Computing Facility
 					 Department of Statistics
 					 UC Berkeley
 					 spector at stat.berkeley.edu
On Fri, 4 Dec 2009, Peng Yu wrote:

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      David Freedman
    

    

      
      #
    

  


  

      
I *think* this is from from 'StatsRUs' - how about

as.data.frame(lapply(df,function(x)rep(x,n)))

hth, david freedman

        
pengyu.ut wrote:

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      William Dunlap
    

    

      
      #
    

  


  

      
Phil Spector's solution is the best way to get your
triplicating job done.

As for why the result of your apply call is a 6 by
4 matrix, read the help file for apply where it talks
about how it 'simplifies' the result.  For FUN's that
do not return a scalar the simplification algorithm
may be surprising if MARGIN!=2: it cbind's the vectors
output by FUN together no matter what the value of
MARGIN was.

R : > m<-matrix(12:1,byrow=TRUE,nrow=4)
R : > m
R :      [,1] [,2] [,3]
R : [1,]   12   11   10
R : [2,]    9    8    7
R : [3,]    6    5    4
R : [4,]    3    2    1
R : > apply(m,1,function(x)x+1000)
R :      [,1] [,2] [,3] [,4]
R : [1,] 1012 1009 1006 1003
R : [2,] 1011 1008 1005 1002
R : [3,] 1010 1007 1004 1001

In addition, calling apply on a data.frame converts
it to a matrix before doing anything and that often
loses or alters information in the data.frame.  You
definitely will not get a data.frame out of apply.

apply has its uses, but it is limited.

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com