CODEWITHSUNDEEP
r
user3355655
user3355655

Reputation: 485

split data frame by sequence series

I have 2 data frames :

p1=rnorm(10)
p2=rnorm(10)
p3=rnorm(10)

pf=data.frame(p1,p2,p3)

pf
           p1          p2         p3
1  -1.4899362  0.44024186  0.5949573
2   1.1727063  0.15883062 -1.4196451
3  -1.4798270  0.65976414 -1.6066772
4  -0.4303878  2.22051966  0.8929259
5  -1.0516386 -1.18394507  0.1481680
6   1.5225863 -0.07395583  1.2270284
7   0.5928281 -0.41635467 -0.7618043
8  -0.2226615 -0.19148234  0.4193754
9   0.7128943  0.06954478 -1.0399434
10  0.7166008  1.15534832  0.7115740

and 

f1<-c(1,2,3,4,5,1,2,3,4,5)
f2<-c(1,2,1,2,3,4,5,6,7,8)
f3<-c(1,2,3,4,5,6,7,1,2,3)

ff<-data.frame(f1,f2,f3)

> ff
   f1 f2 f3
1   1  1  1
2   2  2  2
3   3  1  3
4   4  2  4
5   5  3  5
6   1  4  6
7   2  5  7
8   3  6  1
9   4  7  2
10  5  8  3

How can I split the data frame so that I can have a data frame like this:

I would like to split data pf according ff which contains series

           p1      p1x           p2         p2x           p3            p3x
1  -1.4899362   1.5225863   0.44024186   0.65976414     0.5949573     0.4193754
2   1.1727063   0.5928281   0.15883062   2.22051966    -1.4196451    -1.0399434
3  -1.4798270  -0.2226615     NA        -1.18394507    -1.6066772     0.7115740
4  -0.4303878   0.7128943     NA        -0.07395583     0.8929259        NA
5  -1.0516386   0.7166008     NA        -0.41635467     0.1481680        NA
6       NA          NA        NA        -0.19148234     1.2270284        NA
7       NA          NA        NA         0.06954478    -0.7618043        NA
8       NA          NA        NA         1.15534832        NA            NA

Thanks so much for your help.

Upvotes: 1

Views: 215

Answers (1)

cryo111
cryo111

Reputation: 4474

You haven't provided a seed, but here a possible base R solution:

#find out where to split
split_df=as.data.frame(rbind(matrix(0,ncol=ncol(pf)),
                             apply(diff(as.matrix(ff))!=1,2,cumsum)))
#perform split for each column
split_data=unlist(Map(function(x,y) split(x,y),
                      pf,
                      split_df),recursive=FALSE)
#now bind the data together
max_length=max(sapply(split_data,length))
sapply(split_data,function(x) `[<-`(rep(NA_real_,max_length),
                                    1:length(x),
                                    x))

Upvotes: 3

Related Questions