Re: [R] Testing memory limits in R??
check Memory in R: ?Memory --- On Mon, 7/6/09, Scott Zentz wrote: From: Scott Zentz Subject: [R] Testing memory limits in R?? To: r-help@r-project.org Date: Monday, July 6, 2009, 3:52 PM Hello Everyone, We have recently purchased a server which has 64GB of memory running a 64bit OS and I have compiled R from source with the following config ./configure --prefix=/usr/local/R-2.9.1 --enable-Rshlib --enable-BLAS-shlib --enable-shared --with-readline --with-iconv --with-x --with-tcktk --with-aqua --with-libpng --with-jpeglib and I would like to verify that I can use 55GB-60GB of the 64GB of memory within R. Does anyone know how this is possible? Will R be able to access that amount of memory from a single process? I am not an R user myself but I just wanted to test this before I turned the server over to the researchers.. Thanks! -scz __ R-help@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. [[alternative HTML version deleted]] __ R-help@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.
Re: [R] #INCLUDE
source(_external_file_name) --- On Wed, 7/8/09, Idgarad wrote: From: Idgarad Subject: [R] #INCLUDE To: r-help@r-project.org Date: Wednesday, July 8, 2009, 11:16 AM What is R's equivalent to a C-like #include to incorporate external files. I have a 2k line function that is generated and need to include it at runtime but not manage it as a package (as it changes hourly.) Any ideas? [[alternative HTML version deleted]] __ R-help@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. [[alternative HTML version deleted]] __ R-help@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.
Re: [R] cannot allocate a vector with 1920165909 length
Out of memory? How large is your physical memory? --- On Tue, 7/28/09, zhijie zhang wrote: From: zhijie zhang Subject: [R] cannot allocate a vector with 1920165909 length To: r-h...@stat.math.ethz.ch Date: Tuesday, July 28, 2009, 9:49 PM Dear Rusers, The error for the following was that it cannot allocate the vector of length 1920165909. a <- expand.grid(se1=0:100/100, sp1=0:100/100, se2=0:100/100, sp2=0:100/100, DR=0:100/100) How to solve it? Maybe setwd(dir) can, i am not very sure about it. Any ideas about it? [[alternative HTML version deleted]] __ R-help@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. [[alternative HTML version deleted]] __ R-help@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.
[R] R CMD build/install: wrong Rtools include path is passed to g++
Hi: I am trying to build/install rparallel source package in win32 using Rtools/R CMD. However, R CMD build or install fails. The R CMD build output shows that the path of Rtools/MinGW/include is wrong in g++ -I. How can I pass/configure the correct include path to R CMD? Tried this in both R 2.12 and 2.11 with compatible Rtools and Miktex/chm helper. Neither succeeded. Note, the R/Rtools/MinGW setting works fine if the package doesn't have C/C++ code. I was able to install my own R package which doesn't have C/C++ code. [[alternative HTML version deleted]] __ R-help@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.
Re: [R] R CMD build/install: wrong Rtools include path is passed to g++
Never mind. Found the solution: the package coded the rtools path in Makevars.win. So I was able to compile (but have another problem though). But not sure if there is an environment name for rtools, maybe RTOOLS_HOME ... Thanks. - Forwarded Message From: Andy Zhu Cc: r-help@r-project.org Sent: Mon, December 20, 2010 11:33:31 PM Subject: [R] R CMD build/install: wrong Rtools include path is passed to g++ Hi: I am trying to build/install rparallel source package in win32 using Rtools/R CMD. However, R CMD build or install fails. The R CMD build output shows that the path of Rtools/MinGW/include is wrong in g++ -I. How can I pass/configure the correct include path to R CMD? Tried this in both R 2.12 and 2.11 with compatible Rtools and Miktex/chm helper. Neither succeeded. Note, the R/Rtools/MinGW setting works fine if the package doesn't have C/C++ code. I was able to install my own R package which doesn't have C/C++ code. [[alternative HTML version deleted]] __ R-help@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.
[R] installation of R/parallel package in win32/64
This is to summarize my workaround to install R/parallel in win32/64 boxes. Recently I had problems to install rparallel: 1. The package's Makevars.win coded include fixed path for Rtools 2. package is written in C++; Rtools and R are not intended to run g++ by default. My workaround: 1. Need to install R, Rtools as usual. Note Rtools 2.12 has both 32bit mingw and 64 bit. However, it doesn't include package for g++ and libstdc++. You need to install these 2 packages into rtools from sourceforge first. 2. In rparallel source tree, open Makevars.win: change the PKG_CPPFLAGS and PKG_LIBS variables to point to your correct rtools and rtools/mingw directories. 3. In R_installation/etc: open i386 or win64 (forget the name for 64bit) and open Makeconf file; look for DLLFLAGS += ... and append -static-libstdc++. This flag will cause g++ to statically link in libstdc++. then you can run usual R CMD INSTALL rparallel. Good luck. [[alternative HTML version deleted]] __ R-help@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.
Re: [R] getting time series into r
No attachment. But if your date is in the MMDD then the format syntax is
...format="%Y%m%d").
Best,
--- On Mon, 5/23/11, Heiman, Thomas J. wrote:
From: Heiman, Thomas J.
Subject: [R] getting time series into r
To: "R-help@r-project.org"
Date: Monday, May 23, 2011, 2:32 PM
Hi,
I am trying to get the attached following two timeseries (these are small
subsets of the whole thing) into R so I can merge them using zoo.
tmp <- read.table("baltimore.csv", sep = ",") ##This is timeseries 2
z <- zoo(tmp[, 2:20], as.Date(as.character(tmp[, 1]), format = "%y %m %d"))
tmp1 <- read.table("baltimorefludata.csv", sep = ",") ##This is timeseries 1
z2 <- zoo(tmp[,2], as.Date(tmp[, 1]), format = "%m %d %y"))
R is not recognizing Date or YEARMODA as dates..Any suggestions on what I am
doing wrong or how to fix it would be greatly appreciated!!!
Sincerely,
tom
Thomas Heiman, PhD
Info Systems Eng, Sr
The MITRE Corporation | Center for Enterprise Modernization
Office: 703-983-2951 | thei...@mitre.org
[[alternative HTML version deleted]]
__
R-help@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.
-Inline Attachment Follows-
__
R-help@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.
[[alternative HTML version deleted]]
__
R-help@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.
Re: [R] writing dates to a file
I think if you attach the output file would make clear. I guess it is very
likely your merge has the problem, not the write.table. You may need "by"
parameter in merge based on the date. It also could be date value is converged
to numeric value, in this case you just need to convert it back.
--- On Tue, 5/24/11, Heiman, Thomas J. wrote:
From: Heiman, Thomas J.
Subject: [R] writing dates to a file
To: "R-help@r-project.org"
Date: Tuesday, May 24, 2011, 2:34 PM
Hi,
I have attached the data files to this note. I use this code:
library(zoo)
z1 <- read.zoo("baltimorefludata.txt", format = "%m/%d/%Y", header = TRUE)
z2 <- read.zoo("baltimorew.txt", format = "%Y%m%d", header = TRUE)
z3<-merge(z1,z2)
write.table(z3, "fluweatherdata_baltimore2.txt", sep="\t")
R is writing the other data to the file but not the dates... Is there a way to
also write the dates to the file? Thank you!!
Sincerely,
tom
Thomas Heiman, PhD
Info Systems Eng, Sr
The MITRE Corporation | Center for Enterprise Modernization
Office: 703-983-2951 | thei...@mitre.org
-Inline Attachment Follows-
__
R-help@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.
[[alternative HTML version deleted]]
__
R-help@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.
Re: [R] matrix not working
Dat:
1. you can use as.matrix to convert data.frame to matrix;
2. it is likely that the internal representation of your data.frame may not be
numerical value; matrix can only take on numeric.
--- On Thu, 5/26/11, Dat Mai wrote:
From: Dat Mai
Subject: [R] matrix not working
To: r-help@r-project.org
Date: Thursday, May 26, 2011, 12:24 PM
Hello All,
I'm trying to create a matrix from a dataframe (let's call it df):
..a..b.c.d
a inputs output
b inputs output
c inputs output
d inputs output
e inputs output
The inputs are represented by columns a and b
The outputs are represented by columns c and d, but the only outputs are
those from column d
- some values from column d are NA
- column d was created with the code:
df$d=rank(df$c, na.last="keep")
#--R Code-#
item=unique(df$a)
n=length(list)
r=matrix(data=NA,nrow=n, ncol=n, dimnames=list(PRR1=item, PRR2=item))
for(j in 2:ln)
{
for(i in 1:(j-1))
{
input1=rownames(r)[i]
input2=colnames(r)[j]
q=df[(df$a==input1 & df$b==input2), "d"]
if(length(q)==0)
{
q=df[(df$a==input2 & df$b==input1), "d"]
}
if(length(q)==0)
{
q=NA
}
r[j,i]=q
r[i,j]=q
r[j,j]=q
}
}
The result is a matrix with the appropriate dimensions, but everything is
filled with NA instead of the rankings of the various combinations. I'd like
for the matrix to be filled with the ranking values--what have I done wrong?
--
Best,
Dat Mai
PhD Rotation Student
Albert Einstein College of Medicine
[[alternative HTML version deleted]]
__
R-help@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.
[[alternative HTML version deleted]]
__
R-help@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.
Re: [R] Run R script automatically each morning and email results?
Use cron job scheduler to start r session on your script. Andy From: Sarah Henderson To: R List Sent: Wednesday, June 1, 2011 7:50 PM Subject: [R] Run R script automatically each morning and email results? Greetings to all -- I am hoping that someone can offer some insight into an area where I have very little experience. I have written some R code that generates a 90-day plot of temperature vs. mortality. On whatever day the code is run it will grab up-to-date vital statistics data from our server and combine it with up-to-date temperature data from the web. This is a baby step in automated surveillance of the relationship between temperature and mortality in our area. The next step is to have this code run automatically every morning at 8:00am, and to email the resulting plot to a few key stakeholders. Can anyone recommend how I might accomplish this? Either in R, or some other user-friendlish solution? Many thanks, Sarah [[alternative HTML version deleted]] __ R-help@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. [[alternative HTML version deleted]] __ R-help@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.
[R] R CMD in 2.11.1 error
Hi,
I have a set of user defined package which I have installed in 2.8.1 for quite
some time. I installed 2.11.0 and 2.11.1 a couple of days ago while still
having 2.8.1 version in my system. For the package, package.skeleton works fine
in R 2.11. However, it comes problem when I run Rcmd check/intall. The
intall.out shows below.
I further tested that I reran R-2.8.1/bin/R CMD check in 2.8.1. it still
works. Then I ran and test the doomed function, gzWrite, in R-2.11.1/binRgui,
it also works fine. So, is this the problem in R-2.11.1/bin/Rcmd itself? How
do I fix it? I have MinGW, Perl, etc. installed in my system.
I also attached my function gzWrite below.
Thank you very much.
00check.out:
* install options are ' --no-html'
* installing *source* package 'quantplus' ...
** R
** preparing package for lazy loading
** help
Warning: newline within quoted string at gzWrite.Rd:92
Warning: ./man/gzWrite.Rd:99: unknown macro '\n'
Warning: newline within quoted string at gzWrite.Rd:92
Error in parse_Rd("./man/gzWrite.Rd", encoding = "unknown", ...) :
Unexpected end of input (in " quoted string opened at gzWrite.Rd:99:46)
ERROR: installing Rd objects failed for package 'quantplus'
* removing
'd:/developer/projects/quant+/package_builder/quantplus.Rcheck/quantplus'
the problem function:
**
gzWrite =
# write out a data frame into a gzip file in the format of write.table
function(ds, outdir, outpfx, datepfx, sep='\t', quote=TRUE, row.names=FALSE,
col.names=TRUE, na='NA', append=FALSE) {
if (F) {
ds = ds.price;
outdir = pathPrice;
outpfx = paste(pfx.price,'_',sep='');
datepfx = fdate;
sep = '\t';
quote = TRUE;
row.names = FALSE;
col.names = TRUE;
na = 'NULL';
append = FALSE;
}
file.out = paste(outdir, '\\', outpfx, datepfx, '.gz', sep='');
file.tmp = paste(outdir, '\\', outpfx, datepfx, '.txt',sep='');
write.table(ds, file=file.tmp, sep=sep, quote=quote, row.names=row.names,
col.names=col.names, na=na, append=append);
zz = gzfile(file.out, 'w');
raw = file(file.tmp, 'r');
cat(readLines(raw), file=zz, append=F, sep='\n');
close(zz);
close(raw);
unlink(file.tmp);
invisible(NA);
}
[[alternative HTML version deleted]]
__
R-help@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.
[R] help: error handling in try
First time to post and searched archive for this problem with no clue. My
version is 2.5.1.
Below is a function to check if a given date is a valid date to a given date
function object. It uses try (also tried tryCatch but with same problem). When
given an invalid date, I am hoping try will generate en error message which
would be picked up by the geterrmessage and thus expecting a false result. but
this is not the case when it is inside a function. However, if simply run the
body of function, the ret is indeed a false value. Why are they different?
Many thanks
is.Date=
# adate: a scalar value
# fun: the date function object
# format: the designated date format to date function
function(adate,fun,format) {
ret=NA;
error.old=geterrmessage();
.Internal(seterrmessage('no error'));
try(fun(adate,format),silent=T);
error=geterrmessage();
if (error=='no error') {
ret=T;
} else {
ret=F;
}
.Internal(seterrmessage(error.old));
return(ret);
}
adate='12/2000';
fun=as.Date;
format='%b %Y';
is.Date(adate,fun,format) # returns true which is not correct.
ret=NA;
error.old=geterrmessage();
.Internal(seterrmessage('no error'));
try(fun(adate,format),silent=T);
error=geterrmessage();
if (error=='no error') {
ret=T;
} else {
ret=F;
}
.Internal(seterrmessage(error.old));
ret # shows false which is correct
[[alternative HTML version deleted]]
__
R-help@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.
Re: [R] help: error handling in try
Thanks, Jim:
One of the purpose of this function is to try not to stop the program and
generate T/F result for further process. Actually find fix by myself with
tryCatch (but don't know why try doesn't work). The modified function handles
potential Date and yearmon in Zoo classes.
is.Date=
# adate: a scalar value
# fun: the date function object
# format: the designated date format to date function
function(adate,fun,format) {
ret=NA;
error.old=geterrmessage();
.Internal(seterrmessage('no error'));
temp=tryCatch(fun(adate,format),
error=function(e) {
.Internal(seterrmessage('error'))
}
);
error=geterrmessage();
if (error=='error') {
ret=F;
} else {
if (is.na(temp)) {
ret=F;
} else {
ret=T;
}
}
.Internal(seterrmessage(error.old));
return(ret);
}
adate='12/2000';
fun=as.Date;
format='%b %Y';
is.Date(adate,fun,format); # false
library(zoo);
fun=as.yearmon;
format='%m/%Y';
is.Date(adate,fun,format); # true
Cheer.
--- On Wed, 9/10/08, jim holtman <[EMAIL PROTECTED]> wrote:
From: jim holtman <[EMAIL PROTECTED]>
Subject: Re: [R] help: error handling in try
To: [EMAIL PROTECTED]
Cc: r-help@r-project.org
Date: Wednesday, September 10, 2008, 8:11 AM
Why don't you use 'try' in this fashion:
> f.error <- function(x) if (x == 1) stop('error')
>
> value <- try(f.error(1))
Error in f.error(1) : error
> if (inherits(value, 'try-error')) cat("Got this error:",
value) else print ("no error")
Got this error: Error in f.error(1) : error
> value <- try(f.error(0)) # no error
> if (inherits(value, 'try-error')) cat ("Got this
error:", value) else print("no error")
[1] "no error"
>
>
On Tue, Sep 9, 2008 at 8:56 PM, Andy Zhu <[EMAIL PROTECTED]> wrote:
> First time to post and searched archive for this problem with no clue. My
version is 2.5.1.
>
> Below is a function to check if a given date is a valid date to a given
date function object. It uses try (also tried tryCatch but with same problem).
When given an invalid date, I am hoping try will generate en error message which
would be picked up by the geterrmessage and thus expecting a false result. but
this is not the case when it is inside a function. However, if simply run the
body of function, the ret is indeed a false value. Why are they different?
>
> Many thanks
>
>
> is.Date=
> # adate: a scalar value
> # fun: the date function object
> # format: the designated date format to date function
> function(adate,fun,format) {
> ret=NA;
> error.old=geterrmessage();
> .Internal(seterrmessage('no error'));
> try(fun(adate,format),silent=T);
> error=geterrmessage();
> if (error=='no error') {
> ret=T;
> } else {
> ret=F;
> }
> .Internal(seterrmessage(error.old));
> return(ret);
> }
>
> adate='12/2000';
> fun=as.Date;
> format='%b %Y';
>
> is.Date(adate,fun,format) # returns true which is not correct.
>
> ret=NA;
>
> error.old=geterrmessage();
>
> .Internal(seterrmessage('no error'));
>
> try(fun(adate,format),silent=T);
>
> error=geterrmessage();
>
> if (error=='no error') {
>
> ret=T;
>
> } else {
>
> ret=F;
>
> }
>
> .Internal(seterrmessage(error.old));
>
>
> ret # shows false which is correct
>
>
>
>
>[[alternative HTML version deleted]]
>
>
> __
> R-help@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.
>
>
--
Jim Holtman
Cincinnati, OH
+1 513 646 9390
What is the problem that you are trying to solve?
[[alternative HTML version deleted]]
__
R-help@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.
Re: [R] Woring message in as.yearmon()
Try this modification: as.yearmon(paste(res[,1], res[,2], sep="-"),format="%Y-%m")Andy --- On Wed, 9/10/08, Megh Dal <[EMAIL PROTECTED]> wrote: From: Megh Dal <[EMAIL PROTECTED]> Subject: [R] Woring message in as.yearmon() To: [EMAIL PROTECTED] Date: Wednesday, September 10, 2008, 2:15 PM I have following dataset: > res [,1] [,2] [,3] [1,] 19464 1.27 [2,] 19465 1.27 [3,] 19466 1.27 [4,] 19467 1.27 [5,] 19468 1.52 [6,] 19469 1.52 [7,] 1946 10 1.52 [8,] 1946 11 1.52 [9,] 1946 12 1.62 [10,] 19471 1.62 [11,] 19472 1.62 [12,] 19473 1.62 [13,] 19474 1.87 [14,] 19475 1.87 [15,] 19476 1.87 Now I write following code : > as.yearmon(paste(res[,1], res[,2], sep="-")) [1] "Apr 1946" "May 1946" "Jun 1946" "Jul 1946" "Aug 1946" "Sep 1946" "Oct 1946" "Nov 1946" "Dec 1946" [10] "Jan 1947" "Feb 1947" "Mar 1947" "Apr 1947" "May 1947" "Jun 1947" Warning message: In if (nch == 1) "%Y-%m" else "%Y-%m-%d" : the condition has length > 1 and only the first element will be used Can anyone please tell me why this warning message is coming and what is the remedy? Regards, __ R-help@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. [[alternative HTML version deleted]] __ R-help@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.
Re: [R] help: error handling in try
Hi, Jim:
Thanks again for replying. Sorry to confuse on saying stop the program; I mean
program exits the function early where stop raises the error (this is inside a
function). It seems that try doesn't fit in a function (in my old version) but
tryCatch does (don't use stop in a function to raise error then one needs
another function on top of it).
Not sure why try doesn't work though.
best.
--- On Wed, 9/10/08, jim holtman <[EMAIL PROTECTED]> wrote:
From: jim holtman <[EMAIL PROTECTED]>
Subject: Re: [R] help: error handling in try
To: [EMAIL PROTECTED]
Cc: r-help@r-project.org
Date: Wednesday, September 10, 2008, 9:15 PM
It does not "stop" the program. It generates an "error"
message which
is caught by 'try' from which you can do a recovery. The purpose of
'stop' is to cause the 'try' to catch it and return the error
message
as the result and you can test the class for being 'try-error'.
On Wed, Sep 10, 2008 at 9:11 PM, Andy Zhu <[EMAIL PROTECTED]> wrote:
> Thanks, Jim:
>
> One of the purpose of this function is to try not to stop the program and
generate T/F result for further process. Actually find fix by myself with
tryCatch (but don't know why try doesn't work). The modified function
handles potential Date and yearmon in Zoo classes.
>
> is.Date=
> # adate: a scalar value
> # fun: the date function object
> # format: the designated date format to date function
> function(adate,fun,format) {
> ret=NA;
> error.old=geterrmessage();
> .Internal(seterrmessage('no error'));
> temp=tryCatch(fun(adate,format),
> error=function(e) {
> .Internal(seterrmessage('error'))
> }
> );
> error=geterrmessage();
> if (error=='error') {
> ret=F;
> } else {
> if (is.na(temp)) {
> ret=F;
> } else {
> ret=T;
> }
> }
> .Internal(seterrmessage(error.old));
> return(ret);
> }
>
> adate='12/2000';
> fun=as.Date;
> format='%b %Y';
>
> is.Date(adate,fun,format); # false
>
> library(zoo);
> fun=as.yearmon;
> format='%m/%Y';
>
> is.Date(adate,fun,format); # true
>
> Cheer.
>
> --- On Wed, 9/10/08, jim holtman <[EMAIL PROTECTED]> wrote:
> From: jim holtman <[EMAIL PROTECTED]>
> Subject: Re: [R] help: error handling in try
> To: [EMAIL PROTECTED]
> Cc: r-help@r-project.org
> Date: Wednesday, September 10, 2008, 8:11 AM
>
> Why don't you use 'try' in this fashion:
>
>> f.error <- function(x) if (x == 1) stop('error')
>>
>> value <- try(f.error(1))
> Error in f.error(1) : error
>> if (inherits(value, 'try-error')) cat("Got this
error:",
> value) else print ("no error")
> Got this error: Error in f.error(1) : error
>> value <- try(f.error(0)) # no error
>> if (inherits(value, 'try-error')) cat ("Got this
> error:", value) else print("no error")
> [1] "no error"
>>
>>
>
>
> On Tue, Sep 9, 2008 at 8:56 PM, Andy Zhu <[EMAIL PROTECTED]>
wrote:
>> First time to post and searched archive for this problem with no clue.
My
> version is 2.5.1.
>>
>> Below is a function to check if a given date is a valid date to a
given
> date function object. It uses try (also tried tryCatch but with same
problem).
> When given an invalid date, I am hoping try will generate en error message
which
> would be picked up by the geterrmessage and thus expecting a false result.
but
> this is not the case when it is inside a function. However, if simply run
the
> body of function, the ret is indeed a false value. Why are they different?
>>
>> Many thanks
>>
>>
>> is.Date=
>> # adate: a scalar value
>> # fun: the date function object
>> # format: the designated date format to date function
>> function(adate,fun,format) {
>> ret=NA;
>> error.old=geterrmessage();
>> .Internal(seterrmessage('no error'));
>> try(fun(adate,format),silent=T);
>> error=geterrmessage();
>> if (error=='no error') {
>> ret=T;
>> } else {
>> ret=F;
>> }
>> .Internal(seterrmessage(error.old));
>> return(ret);
>> }
>>
>> adate='12/2000';
>> fun=as.Date;
>> format='%b %Y';
>>
>> is.Date(adate,fun,format) # returns
Re: [R] Passing data among multiple instances
--- On Wed, 2/4/09, Feng Li <840...@gmail.com> wrote: From: Feng Li <840...@gmail.com> Subject: Re: [R] Passing data among multiple instances To: "Warren Young" Cc: r-help@r-project.org Date: Wednesday, February 4, 2009, 10:19 AM On Wed, Feb 4, 2009 at 4:02 PM, Warren Young wrote: > Feng Li wrote: > >> >> I have two R instances running at the same time, >> > > On the same computer, or on different computers? The first trial is only on my single computer with Quad CPU and more than 2G mem. > > > Is the number of Rs likely to change, or will it always be just the two? I am planning to do three tasks at same time. That will be instance A, instance B and instance C. There tasks are more or less the same. But one always depends on others results. > > > Is this a simple one-off problem, or are you breaking the problem up into > pieces so you can throw lots of hardware at it? This is just for one project. But if this is available, later I will try more on this! > > > Is there a simpler way to pass the data in A to B? >> > > Perhaps the simplest option is to write the data structure to a file, using > any of the several R ways to do that. When instance 2 sees that a file is > available, it slurps its contents in and works on it. The hard part is > making the second instance wait until the whole file is written out by the > first. You wouldn't want it to read in half the file then hit the end > because the first process hasn't finished writing out the file. I don't see > any good mechanism in R to fix this. actually, could this be fixed this way: Instance B waits on instance A for a signal file. Instance A first writes the data file as usual (the file you mentioned above). After instance A finishes the data file, it writes a signal file. Instance B checks on signal file existence but it works on the data file. > > A more robust option is to use sockets. This is suitable even within a > single machine. See ?make.socket. This solves the "how do I know when I've > got the full data structure problem" because the second process can just > keep reading until it gets an error indicating that the remote peer closed > the connection. Once you have the data structure in string form, you can > eval() it to get an R object suitable for munching on. Figuring out how to > pass the data might be the hardest part. deparse() might be the easiest > way. > > If you're hoping to scale this up to lots of processes, look into Rmpi. > This provides a very clean way for an R program on one computer to start > slaves on other computers and then pass data to them in native R structures. > Setting up MPI itself is not trivial, however. It's best when you already > have a cluster of computers linked with MPI. > > __ > R-help@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. > -- Feng Li Department of Statistics Stockholm University 106 91 Stockholm, Sweden [[alternative HTML version deleted]] __ R-help@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. [[alternative HTML version deleted]] __ R-help@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.
[R] Q: rbind problem in my code
I have a large matrix which is divided into several pieces, manipulated
individually, and saved into RData in disc. After each pieces are done with
operation, I load them into memory and use rbind to stack them back into
matrix. however, the rbind is only give me the last two pieces. The following
code illustrates the problem: 4 pieces of matrix (1000x1000) is created and
stored in RData; then load them back and rbind them. I would expect mat is of
4000*1000; however, dim(mat) shows (2000 1000).
Where is my code wrong?
Thanks
## define a pathData directory
pathData='C:'\\data';
points=1000;
data.pfx='rnorm';
rname=paste('row',1:points,sep='');
cname=paste('col',1:points,sep='');
for (i in 1:4) {
x=rnorm(points^2);
mat=matrix(data=x,nrow=points,ncol=points);
rname=paste('row',(points*(i-1)+1):(points*i),sep='');
rownames(mat)=rname;
colnames(mat)=cname;
file.rdata=paste(pathData,'\\',data.pfx,'_',i,'.RData',
sep='');
save(list='mat',file=file.rdata,compress=T);
}
mat=NULL;
for (i in 1:4) {
file.rdata=paste(pathData,'\\',data.pfx,'_',i,'.RData',
sep='');
x=load(file.rdata);
mat=rbind(mat,get(x));
}
### code end here
> dim(mat)
[1] 2000 1000
[[alternative HTML version deleted]]
__
R-help@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.
