thr3ads.net - R help - [R] Loop Help [May 2012]

If this information is useful, please help other people find it:
Share via:

bdossman

2012-May-18 23:14 UTC

[R] Loop Help

Hi all,

I am a beginner R user and need some help with a simple loop function.

Currently, I have seven datasets (TOWER1,TOWER2...TOWER7) that are all in
the same format (same # of col and headers). I am trying to add a new column
(factor) to each dataset that simply identifies the dataset. Ultimately, I
would like to merge all 7 datasets and need that column to identify what
rows came from what dataset.

Using the code below, I get the error message "Error in rep(i,
nrow(TOWER.i)) : invalid 'times' argument" but it doesn't make
sense to me
since nrow should give an integer value. Any help will be really
appreciated.

TOWERS<-c("TOWER1","TOWER2","TOWER3","TOWER4","TOWER5","TOWER6","TOWER7")

for(i in 1:7){
	TOWER.i<-TOWERS[i]
	TOWER<-rep(i,nrow(TOWER.i))
	TOWER.i<-cbind(TOWER.i[1:2],TOWER, TOWER.i[2:length(TOWER.i)])
}

--
View this message in context:
http://r.789695.n4.nabble.com/Loop-Help-tp4630555.html
Sent from the R help mailing list archive at Nabble.com.

Simon Knapp

2012-May-19 11:06 UTC

head link

[R] Loop Help

First, what was going wrong:

#this creates a character vector, not an object containing your data sets
TOWERS <-
c("TOWER1","TOWER2","TOWER3","TOWER4","TOWER5","TOWER6","TOWER7")

for(i in 1:7){
    # this creates a variable named "TOWER.i" (not "TOWER.1,
TOWER.2,
..., TOWER.7)
    # which will get overwritten each time through the loop
    TOWER.i <- TOWERS[i]
    # This is what causes your error, as TOWER.i is a string
("TOWER1", "TOWER2", ...), and hence nrow(TOWER.i) return
NULL
 ? ?TOWER <- rep(i,nrow(TOWER.i))
    # if you got to here, you would have got more errors as TOWER.i is
cannot be subsetted
    # (it is a string, not a 'dataset' with columns)
 ? ?TOWER.i <- cbind(TOWER.i[1:2], TOWER, TOWER.i[2:length(TOWER.i)])
}




# Note that to create variables like 'TOWER.1', ..., you would have to
say something like
assign(paste('TOWER', i, sep='.'), TOWERS[[i]])
# inside the loop. However, this is not necessary here and you would
keep needing to
# access the variable with similar ugliness. Here is the way I might
approach this:

TOWERS <- list(TOWER1,TOWER2,TOWER3,TOWER4,TOWER5,TOWER6,TOWER7)

# create a list of things that look like what you wanted TOWERS.i to look like
# (still with column two repeated
TOWERS.modified <- mapply(function(x, i) cbind(x[,1:2], i,
x[,2:ncol(x)]), TOWERS, 1:length(TOWERS), SIMPLIFY=F)

# you can then 'merge' them (though that term would mean something else
to most
# R programmers - see ?merge) with:
TOWERS.combined <- do.call(rbind, TOWERS.modified)




#Now what I would actually do is:
TOWERS.combined <- do.call(rbind, lapply(1:7, function(i) {
    x <- get(paste('TOWER', i, sep=''))
    cbind(x[,1:2], i, x[,2:ncol(x)])
}))



Hopefully that makes sense, if not I can provide more explanation, but
please try ?do.call, ?get, ?lapply and ?mapply first

Simon

On Sat, May 19, 2012 at 9:14 AM, bdossman <bdossman at gmail.com>
wrote:> Hi all,
>
> I am a beginner R user and need some help with a simple loop function.
>
> Currently, I have seven datasets (TOWER1,TOWER2...TOWER7) that are all in
> the same format (same # of col and headers). I am trying to add a new
column
> (factor) to each dataset that simply identifies the dataset. Ultimately, I
> would like to merge all 7 datasets and need that column to identify what
> rows came from what dataset.
>
> Using the code below, I get the error message "Error in rep(i,
> nrow(TOWER.i)) : invalid 'times' argument" but it doesn't
make sense to me
> since nrow should give an integer value. Any help will be really
> appreciated.
>
>
TOWERS<-c("TOWER1","TOWER2","TOWER3","TOWER4","TOWER5","TOWER6","TOWER7")
>
> for(i in 1:7){
> ? ? ? ?TOWER.i<-TOWERS[i]
> ? ? ? ?TOWER<-rep(i,nrow(TOWER.i))
> ? ? ? ?TOWER.i<-cbind(TOWER.i[1:2],TOWER, TOWER.i[2:length(TOWER.i)])
> }
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/Loop-Help-tp4630555.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.

Rui Barradas

2012-May-19 14:03 UTC

head link

[R] Loop Help

Hello,

The error is that you are trying to use the number of rows of a 
character string.
TOWERS[1], TOWERS[2], etc, are not data frames. Use a print statement 
before the line that throws the error to check it.

Your problem can be solved along the lines of what follows.
Note that I've put all data frames in a list, it's better to have them 
kept like that, it makes everything else simpler.


# Create a list with some made up data.frames
towers <- list(TOWER1=data.frame(A=letters[1:4], X=rnorm(4)),
     TOWER2=data.frame(A=LETTERS[1:6], X=runif(6)))
towers

# In your case this is  from TOWER1 to TOWER7
TOWERS <- names(towers)

towers.with.id <- lapply(TOWERS, function(i){
         towers[[ i ]]$Tower <- factor(i)
         towers[[ i ]]})


(Use something other than 'towers.with.id', this is just an example.)


#names(towers.with.id) <- TOWERS  # (*) See below
towers.with.id
do.call(rbind, towers.with.id)


(*) Try to run these last three instructions with the marked line 
commented/uncommented.
It's better to uncomment, maybe after rbind.
You'll later be able to access the list elements with a syntax like


towers.with.id[[ "TOWER2" ]]        # full data.frame 2
towers.with.id[[ TOWERS[2] ]]$A    # just that column
towers.with.id[[ "TOWER2" ]]$A[3]    # third element of that column


If you do.call/rbind before, to solve the rbind-ed data.frame's row 
names use

rownames(result) <- seq.int(nrow(result))

where 'result' is the result of do.call.

Hope this helps,

Rui Barradas

Em 19-05-2012 11:00, r-help-request at r-project.org
escreveu:> Date: Fri, 18 May 2012 16:14:08 -0700 (PDT)
> From: bdossman<bdossman at gmail.com>
> To:r-help at r-project.org
> Subject: [R] Loop Help
> Message-ID:<1337382848213-4630555.post at n4.nabble.com>
> Content-Type: text/plain; charset=us-ascii
>
> Hi all,
>
> I am a beginner R user and need some help with a simple loop function.
>
> Currently, I have seven datasets (TOWER1,TOWER2...TOWER7) that are all in
> the same format (same # of col and headers). I am trying to add a new
column
> (factor) to each dataset that simply identifies the dataset. Ultimately, I
> would like to merge all 7 datasets and need that column to identify what
> rows came from what dataset.
>
> Using the code below, I get the error message "Error in rep(i,
> nrow(TOWER.i)) : invalid 'times' argument" but it doesn't
make sense to me
> since nrow should give an integer value. Any help will be really
> appreciated.
>
>
TOWERS<-c("TOWER1","TOWER2","TOWER3","TOWER4","TOWER5","TOWER6","TOWER7")
>
> for(i in 1:7){
> 	TOWER.i<-TOWERS[i]
> 	TOWER<-rep(i,nrow(TOWER.i))
> 	TOWER.i<-cbind(TOWER.i[1:2],TOWER, TOWER.i[2:length(TOWER.i)])
> }

@ngel

2012-Jun-15 21:41 UTC

head link

[R] Loop Help

Hello all, 
I'm new here and new to R, but I have to admit I'm learning rather fast.
Right now I need to write a loop, and I don't seem to think I'm doing it
properly, judging by the errors I get! What I need to do is to insert my
data (csv-table) into a variable, and take snapshots of my data (it's
longtitudinal) and they 'play' with the snapshot. To give you an
example:

# I Load tnet the library I need
library(tnet)

# I Load my data
net <- read.table("data.txt")

# Check if net is tnet
net <- as.tnet(net, type="longitudinal tnet")

# And slice the data getting new variables for every day
net.101031 <-
as.static.tnet(net[net[,"t"]<=as.POSIXlt("2010-10-31
00:00:00"),])

# Then get the degree of the sliced network
degree_w(net.101031)

# and the mean of each column in the data.frame
apply(net.101031, 2, mean)

Now the problem is that there are way too many "days" to do it
manually and
I need to wrap this part:


net.101031 <-
as.static.tnet(net[net[,"t"]<=as.POSIXlt("2010-10-31
00:00:00"),])
degree_w(net.101031)
apply(net.101031, 2, mean)

in a loop, and print the results. But... 
Can someone please help me...????
Thank you so much in advance!

--
View this message in context:
http://r.789695.n4.nabble.com/Loop-Help-tp4633558.html
Sent from the R help mailing list archive at Nabble.com.

Özgür Asar

2012-Jun-16 03:27 UTC

head link

[R] Loop Help

Hi,

Do you mean printing the results of 

degree_w(net.101031)

and

apply(net.101031, 2, mean) 

in a loop?

see ?print to print the results

and 

?cat as a suporting function.

For instance see the toy example below

for (i in 1:5){
print('------------------------------------')
cat("model",i,"\n")
print(summary(lm(rnorm(10,0,1)~rnorm(10,0,2))))
}

Hope this helps
Ozgur



--
View this message in context:
http://r.789695.n4.nabble.com/Loop-Help-tp4633558p4633572.html
Sent from the R help mailing list archive at Nabble.com.

Özgür Asar

2012-Jun-18 16:43 UTC

head link

[R] Loop Help

@angel,

You can try append option in write.table

for e.g.

for(i in 1:20)
write.table(i,"out.txt",append=T,sep="\t",row.names=F,col.names=F)

This will help you save your all results by writing each results on
subsequent rows.

Hope this helps
Ozgur
>Hello Jim
>thank you so much for your response.
>When I use write.csv, on the csv file is saved only the last loop
>So when I write
>results.matrix <- apply(degree_w(net.static[[i]]), 2, mean)
>write.csv(results.matrix, file = "results.csv")
>instead of getting a 550 line document with all loop results, I only get
the last one!>Could you please help me?
>Thank you so much in advance 
--
View this message in context:
http://r.789695.n4.nabble.com/Loop-Help-tp4633558p4633722.html
Sent from the R help mailing list archive at Nabble.com.

Reasonably Related Threads

Search for more reasonably related threads

R help - May 2012 - Loop Help

[R] Loop Help

[R] Loop Help

[R] Loop Help

[R] Loop Help

[R] Loop Help

[R] Loop Help

Reasonably Related Threads