apply function over same column of all objects in a list

I'm a little confused on your data structure -- can you use dput() as
described here [1] to give a small reproducible example?

http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example

Michael

Thank you!!! But I realise I've simplified my data to the point that your
solution doesn't actually work -- not your fault, mine! My list is actually
more complicated than what I presented it to be; it's not composed of
numerical matrices but of lists, each being composed of 7 columns, the first
two of each (the ones that I am interested in, x and y) being numerical.
This is what my list really looks like (truncated):

summary(mylist)

      Length Class   Mode
moh6  7      density list
moh7  7      density list
moh8  7      density list
...
etc.

summary(mylist[[3]]) # I've taken number 3 as an example, but they're all
the same

          Length Class  Mode
x         512    -none- numeric
y         512    -none- numeric
bw          1    -none- numeric
n           1    -none- numeric
call        2    -none- call
data.name   1    -none- character
has.na      1    -none- logical

Any suggestion how to get the max of column y for each sub-list (moh6, moh7,
moh8 etc.), and to the max of all these individual maxes? By the way,

lapply(list1,FUN=function(x)x[,2][which.max(x[,2])])

gives me the individual maxes (for a list composed of numerical matrices),
but how do I get to the overall max?

Possibly

max(unlist(list1))

if I understand you correctly.

Many thanks ...



--
View this message in context: http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4638705.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
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.

Michael, here is the data I am working with. I've shortened it without, I
trust, changing its nature. I am not using any package other than what is
pre-installed. Many thanks, Gail

dput(list.example)

list

(structure(list(

x = c(379.455895016957, 380.116796110287, 
380.777697203618, 381.438598296949, 382.099499390279, 382.76040048361, 
383.421301576941, 384.082202670271, 384.743103763602, 385.404004856933, 
386.064905950264, 386.725807043594), 

y = c(7.66474496376218e-06, 
6.43440919959401e-06, 5.36894293042274e-06, 4.45194069157015e-06, 
3.66781132570054e-06, 3.00181397191447e-06, 2.45773671502347e-06, 
2.00342333333225e-06, 1.62212866305772e-06, 1.3043460449395e-06, 
1.04141214324248e-06, 8.26430099678339e-07), 

bw = 9.02854660253144, 

n = 1102L,

call = density.default(x = read.table(paste(locuteur, 
        i, sep = ""))$V1[read.table(paste(locuteur, i, sep = ""))$V1 != 
        0]),

data.name = c("read.table(paste(locuteur, i, sep =
\"\"))$V1[read.table(paste(locuteur, ", 
    "    i, sep = \"\"))$V1 != 0]"), has.na = FALSE),

.Names = c("x", 
"y", "bw", "n", "call", "data.name", "has.na"),

class = "density"),

structure(list(x = c(372.576546447762, 373.241257067042, 
    373.905967686322, 374.570678305602, 375.235388924882, 375.900099544162, 
    376.564810163442, 377.229520782722, 377.894231402002, 378.558942021282, 
    379.223652640562, 379.888363259842), y = c(1.60229850505427e-05, 
    1.30822330219271e-05, 1.05973406669671e-05, 8.51514876035057e-06, 
    6.78543779138952e-06, 5.41596769065147e-06, 4.28876937728122e-06, 
    3.36815237991786e-06, 2.62273082547721e-06, 2.02454969122432e-06, 
    1.54894049348427e-06, 1.18505840947482e-06), bw = 7.90353278894747, 
        n = 1409L, call = density.default(x = read.table(paste(locuteur, 
            i, sep = ""))$V1[read.table(paste(locuteur, i, sep = ""))$V1 != 
            0]), data.name = c("read.table(paste(locuteur, i, sep =
\"\"))$V1[read.table(paste(locuteur, ", 
        "    i, sep = \"\"))$V1 != 0]"), has.na = FALSE), .Names = c("x", 
    "y", "bw", "n", "call", "data.name", "has.na"), class = "density"),

    structure(list(x = c(383.473231160603, 384.169713215726, 
    384.866195270849, 385.562677325973, 386.259159381096, 386.955641436219, 
    387.652123491343, 388.348605546466, 389.045087601589, 389.741569656713, 
    390.438051711836, 391.134533766959), y = c(2.40798446492758e-06, 
    2.04205952412882e-06, 1.72782287870817e-06, 1.45399002130161e-06, 
    1.21665858527511e-06, 1.01213493683702e-06, 8.37503296444613e-07, 
    6.94344355906326e-07, 5.72200769661348e-07, 4.6862434236517e-07, 
    3.81353514437358e-07, 3.08309463721622e-07), bw = 10.0108886969865, 
        n = 1522L, call = density.default(x = read.table(paste(locuteur, 
            i, sep = ""))$V1[read.table(paste(locuteur, i, sep = ""))$V1 != 
            0]), data.name = c("read.table(paste(locuteur, i, sep =
\"\"))$V1[read.table(paste(locuteur, ", 
        "    i, sep = \"\"))$V1 != 0]"), has.na = FALSE), .Names = c("x", 
    "y", "bw", "n", "call", "data.name", "has.na"), class = "density"),

    NULL, NULL, NULL, NULL, NULL, NULL)




--
View this message in context: http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4638865.html
Sent from the R help mailing list archive at Nabble.com.

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/476b5a1e/attachment.pl>

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/a56c76d0/attachment.pl>

Hi,

my dataset is the result of the function "density" on another set of data.
It refers to that data in its variable "call", though since all the results
are actually reproduced (except that I've removed all rows bar 10), I am not
sure why R still needs it. But I've understood now why your code doesn't
work on my data: your sublists are still just dataframes, whereas my
sublists contain all sorts of vectors (2 numerical vectors and 5 other
things). I've changed your data to make it look a bit more like mine, and
you'll see that the code doesn't work. 

mango <- list(data.frame(coconut=1:6), y=c(4,5,3,2,1,8),
bw=c(4,18,12,3,4,9), nn=paste0("a",1:6))
banana <- list(data.frame(coconut=c(1,2,18,16,15)), y=c(4,5,9,2,1),
bw=c(4,18,22,3,4), nn=paste0("a",1:5))
pineapple <- list(data.frame(coconut=c(4,6,9)), y=c(8,24,12),
bw=c(14,31,36), nn=paste0("a",1:3))
list4 <- list(mango, banana, pineapple)

b <- list()

 for(i in 1:3){
 b[[i]] <- list()
 b[[i]] <- lapply (list4[[i]]$coconut, FUN=function(x) x[which.max(x)])
 }

b
summary( b)

What I want to end up with is the following:

           coconut
mango       6
banana     18
pineapple   9
(that's for the individual sublists)

and 
           coconut
                18
(that's for the list as a whole).

Hope this is becoming a bit clearer ...



--
View this message in context: http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4638876.html
Sent from the R help mailing list archive at Nabble.com.

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/73607b6a/attachment.pl>

Well but the whole point was to be able to plug in a single command and get
the result for all sublists. If I have to create a new list and type in each
sublist and the column I'm interested in, that completely defeats the
purpose, I might as well just ask for max(sublist$coconut) for each sublist,
that would be just as fast ...



--
View this message in context: http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4638888.html
Sent from the R help mailing list archive at Nabble.com.

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/e578ecb1/attachment.pl>

I still try to figure out, what you are finally asking for, but maybe

res<-sapply(list4,function(x)max(x[[1]]$coconut))
names(res)<-c("mango", "banana", "pineapple")
res
max(res)

is worth a try?

Especially i did not get the point of doing something like
x[which.max(x)] because this is in any case just max(x)

if you would like to plug in different functions or want to select
different columns,

fun<-max # or min, median etc
sapply(lapply(list4,"[[",1),fun)

could be a start, where lapply(list4,"[[",1) extracts the first object
in each sublist - which is in your case a data.frame with just one
column ('coconut'), where 'fun' can be applied to. So to address
'coconut' explicitly, just nest this lapply approach:

sapply(lapply(lapply(list4,"[[",1),"[[","coconut"),fun)
#or use an anonymous extractor function
sapply(lapply(list4,function(x)x[[1]][["coconut"]]),fun)

cheers

Am 02.08.2012 16:09, schrieb gail:

Hi,

my dataset is the result of the function "density" on another set of data.
It refers to that data in its variable "call", though since all the results
are actually reproduced (except that I've removed all rows bar 10), I am not
sure why R still needs it. But I've understood now why your code doesn't
work on my data: your sublists are still just dataframes, whereas my
sublists contain all sorts of vectors (2 numerical vectors and 5 other
things). I've changed your data to make it look a bit more like mine, and
you'll see that the code doesn't work. 

mango <- list(data.frame(coconut=1:6), y=c(4,5,3,2,1,8),
bw=c(4,18,12,3,4,9), nn=paste0("a",1:6))
banana <- list(data.frame(coconut=c(1,2,18,16,15)), y=c(4,5,9,2,1),
bw=c(4,18,22,3,4), nn=paste0("a",1:5))
pineapple <- list(data.frame(coconut=c(4,6,9)), y=c(8,24,12),
bw=c(14,31,36), nn=paste0("a",1:3))
list4 <- list(mango, banana, pineapple)

b <- list()

 for(i in 1:3){
 b[[i]] <- list()
 b[[i]] <- lapply (list4[[i]]$coconut, FUN=function(x) x[which.max(x)])
 }

b
summary( b)

What I want to end up with is the following:

           coconut
mango       6
banana     18
pineapple   9
(that's for the individual sublists)

and 
           coconut
                18
(that's for the list as a whole).

Hope this is becoming a bit clearer ...



--
View this message in context: http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4638876.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
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.

Eik Vettorazzi
Institut f?r Medizinische Biometrie und Epidemiologie
Universit?tsklinikum Hamburg-Eppendorf

Martinistr. 52
20246 Hamburg

T ++49/40/7410-58243
F ++49/40/7410-57790

--
Pflichtangaben gem?? Gesetz ?ber elektronische Handelsregister und Genossenschaftsregister sowie das Unternehmensregister (EHUG):

Universit?tsklinikum Hamburg-Eppendorf; K?rperschaft des ?ffentlichen Rechts; Gerichtsstand: Hamburg

Vorstandsmitglieder: Prof. Dr. Guido Sauter (Vertreter des Vorsitzenden), Dr. Alexander Kirstein, Joachim Pr?l?, Prof. Dr. Dr. Uwe Koch-Gromus

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/1e1a75e1/attachment.pl>

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/a0c00b2e/attachment.pl>

Hi Arun,
if you name the data.frame in the list, e.g.
mango <- list( y=c(4,5,3,2,1,8), bw=c(4,18,12,3,4,9),
nn=paste0("a",1:6),target=data.frame(coconut=1:6))
banana <- list(target=data.frame(coconut=c(1,2,18,16,15)), y=c(4,5,9,2,1),
bw=c(4,18,22,3,4), nn=paste0("a",1:5))
pineapple <- list(target=data.frame(coconut=c(4,6,9)), y=c(8,24,12),
bw=c(14,31,36), nn=paste0("a",1:3))
list6<-list(mango,banana,pineapple)

then
sapply(lapply(list5,"[[","target"),max)

still works.

Regards

Am 02.08.2012 20:21, schrieb arun kirshna [via R]:

Hi Elk,

I tried to test with another case where coconut is in different position in
the sublist.
mango <- list( y=c(4,5,3,2,1,8), bw=c(4,18,12,3,4,9),
nn=paste0("a",1:6),data.frame(coconut=1:6))
banana <- list(data.frame(coconut=c(1,2,18,16,15)), y=c(4,5,9,2,1),
bw=c(4,18,22,3,4), nn=paste0("a",1:5))
pineapple <- list(data.frame(coconut=c(4,6,9)), y=c(8,24,12),
bw=c(14,31,36), nn=paste0("a",1:3)) 
list5<-list(mango,banana,pineapple)
#Your code
fun<-max # or min, median etc
sapply(lapply(list5,"[[",1),fun) 
[1]  8 18  9
# Here, first number should be 6, instead it did the max of y from mango.
#Then, I tried second solution of yours
sapply(lapply(lapply(list5,"[[",1),"[[","coconut"),fun) 
#Error in FUN(X[[1L]], ...) : subscript out of bounds
#Third solution
 sapply(lapply(list5,function(x)x[[1]][["coconut"]]),fun)
#Error in x[[1]][["coconut"]] : subscript out of bounds

#I tried with another way to come up with the result (not as elegant)
b3<-do.call("c",list5)
names(b3)[4]<-"mango"
 names(b3)[5]<-"banana"
 names(b3)[9]<-"pineapple"
 b4<-data.frame(key=names(b3),value=unlist(lapply(b3,FUN=function(x)
max(x))))
 key1<-c("mango","banana","pineapple")
 b5<-subset(b4, b4$key%in% key1)
 b5
#        key value
#4     mango     6
#5    banana    18
#9 pineapple     9
b6<-within(b5,{value<-as.numeric(as.character(value))})
max(b6$value)
#[1] 18
A.K.

______________________________________
If you reply to this email, your message will be added to the discussion below:
http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4638926.html
This email was sent by arun kirshna (via Nabble)

	[[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list
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.

Eik Vettorazzi
Institut f?r Medizinische Biometrie und Epidemiologie
Universit?tsklinikum Hamburg-Eppendorf

Martinistr. 52
20246 Hamburg

T ++49/40/7410-58243
F ++49/40/7410-57790

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120802/c6fb1e96/attachment.pl>

Excellent, thank you to both of you!!!
Gail



--
View this message in context: http://r.789695.n4.nabble.com/apply-function-over-same-column-of-all-objects-in-a-list-tp4638681p4639009.html
Sent from the R help mailing list archive at Nabble.com.