[R] Identify row indices corresponding to each distinct row of a matrix

William Dunlap wdunl@p @ending from tibco@com
Thu Nov 8 23:16:41 CET 2018 

One way, rather clumsy, is to convert your data.frame in a character vector
or list. via an invertible tranformation, and use match on it.  E.g.,

> tmp <- do.call(paste, c(list(sep="\001"), unname(C))) # convert to
character
> # or tmp <- split(C, seq_len(nrow(C))) # convert to list of its rows
> g <- match(tmp, unique(tmp))
> g
 [1] 1 2 3 4 5 6 7 3 8 4 5 2 1 8 1 3 5 7 2 6 7 8 4 6
> split(seq_along(g), g)
$`1`
[1]  1 13 15

$`2`
[1]  2 12 19
...
$`8`
[1]  9 14 22

Both ways work for nice enough inputs, but converting to text can cause
problems if the 'sep' is in any of input text and match() on lists of lists
used
to have problems when the inner lists were big.


Bill Dunlap
TIBCO Software
wdunlap tibco.com

On Thu, Nov 8, 2018 at 1:42 PM, li li <hannah.hlx using gmail.com> wrote:

> Thanks to all the reply. I will try to use plain text in the future.
> One question regarding using "which( ! duplicated( m, MARGIN=1 ) )".
> This seems to return the fist row indices corresponding to the distinct
> rows but it does not give all the row indices
> corresponding to each of the distinct rows. For example, in the my example
> below, rows 1, 13 15 are all (1,9).
> Thanks.
>   Hanna
> > A <- matrix(c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16),8,2)
> > B <- rbind(A,A,A)
> > C <- as.data.frame(B[sample(nrow(B)),])
> > C
>    V1 V2
> 1   1  9
> 2   2 10
> 3   3 11
> 4   5 13
> 5   7 15
> 6   6 14
> 7   4 12
> 8   3 11
> 9   8 16
> 10  5 13
> 11  7 15
> 12  2 10
> 13  1  9
> 14  8 16
> 15  1  9
> 16  3 11
> 17  7 15
> 18  4 12
> 19  2 10
> 20  6 14
> 21  4 12
> 22  8 16
> 23  5 13
> 24  6 14
> > T <- unique(C)
> > T
>   V1 V2
> 1  1  9
> 2  2 10
> 3  3 11
> 4  5 13
> 5  7 15
> 6  6 14
> 7  4 12
> 9  8 16
> >
> > i <- 1
> > which(C[,1]==T[i,1]& C[,2]==T[i,2])
> [1]  1 13 15
>
>
> Bert Gunter <bgunter.4567 using gmail.com> 于2018年11月8日周四 上午10:43写道：
>
> > Yes -- much better than mine. I didn't know about the MARGIN argument of
> > duplicated().
> >
> > -- Bert
> >
> >
> > On Wed, Nov 7, 2018 at 10:32 PM Jeff Newmiller <jdnewmil using dcn.davis.ca.us
> >
> > wrote:
> >
> >> Perhaps
> >>
> >> which( ! duplicated( m, MARGIN=1 ) )
> >>
> >> ? (untested)
> >>
> >> On November 7, 2018 9:20:57 PM PST, Bert Gunter <bgunter.4567 using gmail.com
> >
> >> wrote:
> >> >A mess -- due to your continued use of html formatting.
> >> >
> >> >But something like this may do what you want (hard to tell with the
> >> >mess):
> >> >
> >> >> m <- matrix(1:16,nrow=8)[rep(1:8,2),]
> >> >> m
> >> >      [,1] [,2]
> >> > [1,]    1    9
> >> > [2,]    2   10
> >> > [3,]    3   11
> >> > [4,]    4   12
> >> > [5,]    5   13
> >> > [6,]    6   14
> >> > [7,]    7   15
> >> > [8,]    8   16
> >> > [9,]    1    9
> >> >[10,]    2   10
> >> >[11,]    3   11
> >> >[12,]    4   12
> >> >[13,]    5   13
> >> >[14,]    6   14
> >> >[15,]    7   15
> >> >[16,]    8   16
> >> >> vec <- apply(m,1,paste,collapse="-") ## converts rows into character
> >> >vector
> >> >> vec
> >> >[1] "1-9"  "2-10" "3-11" "4-12" "5-13" "6-14" "7-15" "8-16" "1-9"
> >> >"2-10"
> >> >"3-11" "4-12" "5-13" "6-14"
> >> >[15] "7-15" "8-16"
> >> >> ## Then maybe:
> >> >> tapply(seq_along(vec),vec, I)
> >> >$`1-9`
> >> >[1] 1 9
> >> >
> >> >$`2-10`
> >> >[1]  2 10
> >> >
> >> >$`3-11`
> >> >[1]  3 11
> >> >
> >> >$`4-12`
> >> >[1]  4 12
> >> >
> >> >$`5-13`
> >> >[1]  5 13
> >> >
> >> >$`6-14`
> >> >[1]  6 14
> >> >
> >> >$`7-15`
> >> >[1]  7 15
> >> >
> >> >$`8-16`
> >> >[1]  8 16
> >> >
> >> >> ## gives the row numbers for each unique row
> >> >
> >> >There may well be slicker ways to do this -- if this is actually what
> >> >you
> >> >want to do.
> >> >
> >> >-- Bert
> >> >
> >> >
> >> >
> >> >On Wed, Nov 7, 2018 at 7:56 PM li li <hannah.hlx using gmail.com> wrote:
> >> >
> >> >> Hi all,
> >> >>    I use the following example to illustrate my question. As you can
> >> >see,
> >> >> in matrix C some rows are repeated and I would like to find the
> >> >indices of
> >> >> the rows corresponding to each of the distinct rows.
> >> >>   For example, for the row c(1,9), I have used the "which" function
> >> >to
> >> >> identify the row indices corresponding to c(1,9). Using this
> >> >approach, in
> >> >> order to cover all distinct rows, I need to use a for loop.
> >> >>    I am wondering whether there is an easier way where a for loop can
> >> >be
> >> >> avoided?
> >> >>    Thanks very much!
> >> >>       Hanna
> >> >>
> >> >>
> >> >>
> >> >> > A <- matrix(c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16),8,2)> B <-
> >> >> rbind(A,A,A)> C <- as.data.frame(B[sample(nrow(B)),])> C   V1 V2
> >> >> 1   1  9
> >> >> 2   2 10
> >> >> 3   3 11
> >> >> 4   5 13
> >> >> 5   7 15
> >> >> 6   6 14
> >> >> 7   4 12
> >> >> 8   3 11
> >> >> 9   8 16
> >> >> 10  5 13
> >> >> 11  7 15
> >> >> 12  2 10
> >> >> 13  1  9
> >> >> 14  8 16
> >> >> 15  1  9
> >> >> 16  3 11
> >> >> 17  7 15
> >> >> 18  4 12
> >> >> 19  2 10
> >> >> 20  6 14
> >> >> 21  4 12
> >> >> 22  8 16
> >> >> 23  5 13
> >> >> 24  6 14> T <- unique(C)> T  V1 V2
> >> >> 1  1  9
> >> >> 2  2 10
> >> >> 3  3 11
> >> >> 4  5 13
> >> >> 5  7 15
> >> >> 6  6 14
> >> >> 7  4 12
> >> >> 9  8 16> > i <- 1                    > which(C[,1]==T[i,1]&
> >> >> C[,2]==T[i,2])[1]  1 13 15
> >> >>
> >> >>         [[alternative HTML version deleted]]
> >> >>
> >> >> ______________________________________________
> >> >> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> >> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> >> PLEASE do read the posting guide
> >> >> http://www.R-project.org/posting-guide.html
> >> >> and provide commented, minimal, self-contained, reproducible code.
> >> >>
> >> >
> >> >       [[alternative HTML version deleted]]
> >> >
> >> >______________________________________________
> >> >R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> >> >https://stat.ethz.ch/mailman/listinfo/r-help
> >> >PLEASE do read the posting guide
> >> >http://www.R-project.org/posting-guide.html
> >> >and provide commented, minimal, self-contained, reproducible code.
> >>
> >> --
> >> Sent from my phone. Please excuse my brevity.
> >>
> >
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/
> posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

	[[alternative HTML version deleted]]

More information about the R-help mailing list