R help - Jun 2008 - re sultant column names from reshape::cast, with a fun.aggregate vector

try this: 

scores.melt = data.frame(grade = floor(runif(100, 1,10)), variable
'score', value = rnorm(100));

cast(scores.melt, grade ~ variable, fun.aggregate = c(mean, length))

it has the nice column names of:

  grade  score_mean score_length
1     1  0.08788535            8
2     2  0.16720313           15
3     3  0.41046299            7
4     4  0.13928356           13
...

but now try this:

cast(scores.melt, grade ~ variable, fun.aggregate = c(mean, function(x)
sum(x < 0)))

and you get a huge mess:

  grade  score_mean score_function.x..sum.x...0.
1     1  0.08788535                            4
2     2  0.16720313                            6
3     3  0.41046299                            2
4     4  0.13928356                            5

I would think that something like this would fix it up, but no dice:

 cast(scores.melt, grade ~ variable, fun.aggregate = c(mean, num.neg function(x)
sum(x < 0)))

that is, why not look at names(fun.aggregate)?  or am I missing something?

thanks,
Mike


I would think that 
-- 
View this message in context:
http://www.nabble.com/resultant-column-names-from-reshape%3A%3Acast%2C-with-a-fun.aggregate-vector-tp17910885p17910885.html
Sent from the R help mailing list archive at Nabble.com.

> I would think that something like this would fix it up, but no dice:
>
>  cast(scores.melt, grade ~ variable, fun.aggregate = c(mean, num.neg >
function(x) sum(x < 0)))
>
> that is, why not look at names(fun.aggregate)?  or am I missing something?
Yes, that's a bug in each (the function which turns a vector of
functions into a function that returns a named vector of outputs).
I've added a note to my to do.  In the meantime, you can do:

num.neg <- function(x) sum(x < 0)
cast(scores.melt, grade ~ variable, fun.aggregate = c(mean, num.neg))

Hadley


-- 
http://had.co.nz/