Here're a couple alternatives if you want to use the index instead of
the variable name:
# Reproducible data frame
a1 - 1:15
a2 - letters[1:15]
a3 - LETTERS[1:15]
a4 - 15:1
a5 - letters[15:1]
df - data.frame(a1, a2, a3, a4, a5)
df
# a1 a2 a3 a4 a5
# 1 1 a A 15 o
# 2 2 b B 14 n
# 3
Is this what you're going for?
factor(values, levels=mylevels[[1]], labels=mylevels[[2]])
[1] a b e e j
Levels: a b c d e f g h i j
On 2014-03-28 16:38, Jonathan Greenberg wrote:
R-helpers:
Hopefully this is an easy one. Given a lookup table:
mylevels -
If you rename two or more levels with the same string, R collapses them.
Using gsub like you suggest works well. If there are more than the two
characters causing problems you may need to expand the pattern that's
being matched. With the example you sent, this works:
df - structure(c(6L, 4L,
Out of curiosity, what's the reason for not using plyr?
If it has to do with installing the package, here's how the rbind.fill
function is implemented in plyr. It may help show you what you'd like to
do:
plyr::rbind.fill
function (...)
{
dfs - list(...)
if (length(dfs) == 0)
dplyr's group_by and mutate can create those columns for you:
var1 - c(a,b,c,a,b,c,a,b,c)
var2 - c(X,X,X,Y,Y,Y,Z,Z,Z)
var3 - c(1,2,2,5,2,6,7,4,4)
df - data.frame(var1,var2,var3)
dt - tbl_df(df)
dt %.%
group_by(var2) %.%
mutate(
div = var3[var1 == c],
result_calc = var3/div
)
Does this help?
set.seed(500)
mtx - matrix(round(runif(100,0,1)),
ncol=5,
dimnames=list(Subject=1:20,
Group=c('A','B','C','D','E')))
groupCombo - apply(mtx, 1, function(x){
x - paste(
names(x)[as.logical(x)],
collapse='+'
);
return(x)
})
The reshape2 package may be what you're looking for:
http://cran.r-project.org/web/packages/reshape2/index.html
install.packages(reshape2)
reshape2::melt(df)
On 2014-03-24 12:02, Tham Tran wrote:
Dears R Users,
I have another question on function stack.
Given a data frame like:
I sent that last email too quickly, the duplicate column names cause a
problem when melting. Transposing the data frame first gets around it:
reshape2::melt(t(df))[c('Var1', 'value')]
Var1 value
1a 3
2b 2
3a 9
4b 6
5a 5
6b 8
7a 1
8b
Is this what you're trying to do?
mtrx - structure(c(BB, AB, BB, BB, BB, BB, AA, BB,
BB,
BB, BB, BB, AB, AB, BB, AA, BB,
BB, BB, BB,
BB, AB, BB, BB, BB), .Dim = c(5L, 5L),
.Dimnames = list(
c(rs1, rs2, rs3, rs4, rs5), c(LG_1,
9 matches
Mail list logo
