R help - Sep 2011 - Counting similar rows

Startsituation: 

structure(c(1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 1, 
0, 1, 1), .Dim = 4:5, .Dimnames = structure(list(subject = c("s1", 
"s2", "s3", "s4"), class = c("c1",
"c2", "c3", "c4", "c5")), .Names
c("subject",
"class")), class = c("xtabs", "table"), call =
xtabs(formula = ~subject +
    class, data = ia))

I want a count of subjects that match the same classes in subject_cnt and a
count of the number of classes in class_cnt The result of this example
should be:

structure(list(subject_cnt = c(2L, 1L, 1L), class_cnt = c(3L, 
2L, 3L), c1 = c(1L, 1L, 0L), c2 = c(0L, 1L, 0L), c3 = c(1L, 0L, 
1L), c4 = c(0L, 0L, 1L), c5 = c(1L, 0L, 1L)), .Names =
c("subject_cnt",
"class_cnt", "c1", "c2", "c3",
"c4", "c5"), class = "data.frame", row.names
= c(NA, 
-3L))

How can I achieve this in R, without complicated loops? 

PS. Note that the number of classes and subjects are in real quite big.

Cheers,


Lars



--
View this message in context:
http://r.789695.n4.nabble.com/Counting-similar-rows-tp3847051p3847051.html
Sent from the R help mailing list archive at Nabble.com.

Metronome123 wrote on 09/27/2011 07:24:50 AM:
> 
> Startsituation: 
> 
> structure(c(1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 1, 
> 0, 1, 1), .Dim = 4:5, .Dimnames = structure(list(subject =
c("s1",
> "s2", "s3", "s4"), class = c("c1",
"c2", "c3", "c4", "c5")), .Names >
c("subject",
> "class")), class = c("xtabs", "table"), call
= xtabs(formula = ~subject
+ 
>     class, data = ia))
> 
> I want a count of subjects that match the same classes in subject_cnt 
and a
> count of the number of classes in class_cnt The result of this example
> should be:
> 
> structure(list(subject_cnt = c(2L, 1L, 1L), class_cnt = c(3L, 
> 2L, 3L), c1 = c(1L, 1L, 0L), c2 = c(0L, 1L, 0L), c3 = c(1L, 0L, 
> 1L), c4 = c(0L, 0L, 1L), c5 = c(1L, 0L, 1L)), .Names =
c("subject_cnt",
> "class_cnt", "c1", "c2", "c3",
"c4", "c5"), class = "data.frame",
row.names
> = c(NA, 
> -3L))
> 
> How can I achieve this in R, without complicated loops? 
> 
> PS. Note that the number of classes and subjects are in real quite big.
> 
> Cheers,
> 
> 
> Lars

Try this:

xt <- structure(c(1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 1, 0, 1, 
1), 
        .Dim = 4:5, 
        .Dimnames = structure(list(subject = c("s1", "s2",
"s3", "s4"),
        class = c("c1", "c2", "c3",
"c4", "c5")),
        .Names = c("subject", "class")), 
        class = c("xtabs", "table")) 
df <- as.data.frame(unclass(xt))

dfu <- unique(df)
class_cnt <- apply(dfu, 1, sum)
subject_cnt <- tabulate(match(apply(df, 1, paste, collapse="-"), 
        apply(dfu, 1, paste, collapse="-")))

data.frame(subject_cnt, class_cnt, dfu)


Jean
	[[alternative HTML version deleted]]