Re: [Rd] mean relative differences from all.equal() (PR#9276)
On Thu, 2006-10-05 at 03:10 +0200, [EMAIL PROTECTED] wrote: Full_Name: Brad Christoffersen Version: 2.3.1 OS: Windows XP Submission from: (NULL) (128.196.193.132) Why is the difference between two numbers so different from the mean relative difference output from the all.equal() function? Is this an artifact of the way R stores numerics? I could not find this problem as I searched through the submitted bugs. But I am brand new to R so I apologize if there is something obvious I'm missing here. rm(list=ls(all=TRUE)) ## Remove all objects that could hinder w/ consistent output a - 204 b - 203.9792 all.equal(a,b) [1] Mean relative difference: 0.0001019608 a - b [1] 0.0208 Read the Details section of ?all.equal, which states: Numerical comparisons for scale = NULL (the default) are done by first computing the mean absolute difference of the two numerical vectors. If this is smaller than tolerance or not finite, absolute differences are used, otherwise relative differences scaled by the mean absolute difference. If scale is positive, absolute comparisons are made after scaling (dividing) by scale Thus on R version 2.4.0 (2006-10-03): all.equal(a, b, scale = 1) [1] Mean scaled difference: 0.0208 Please do not report doubts about behavior as bugs. Simply post a query on r-help first. If it is a bug, somebody will confirm it and you can then report it as such. BTW, time to upgrade...Go Wildcats! HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] mean relative differences from all.equal() (PR#9276)
On Wed, 2006-10-04 at 20:22 -0500, Marc Schwartz wrote: On Thu, 2006-10-05 at 03:10 +0200, [EMAIL PROTECTED] wrote: Full_Name: Brad Christoffersen Version: 2.3.1 OS: Windows XP Submission from: (NULL) (128.196.193.132) Why is the difference between two numbers so different from the mean relative difference output from the all.equal() function? Is this an artifact of the way R stores numerics? I could not find this problem as I searched through the submitted bugs. But I am brand new to R so I apologize if there is something obvious I'm missing here. rm(list=ls(all=TRUE)) ## Remove all objects that could hinder w/ consistent output a - 204 b - 203.9792 all.equal(a,b) [1] Mean relative difference: 0.0001019608 a - b [1] 0.0208 Read the Details section of ?all.equal, which states: Numerical comparisons for scale = NULL (the default) are done by first computing the mean absolute difference of the two numerical vectors. If this is smaller than tolerance or not finite, absolute differences are used, otherwise relative differences scaled by the mean absolute difference. If scale is positive, absolute comparisons are made after scaling (dividing) by scale Thus on R version 2.4.0 (2006-10-03): all.equal(a, b, scale = 1) [1] Mean scaled difference: 0.0208 Please do not report doubts about behavior as bugs. Simply post a query on r-help first. If it is a bug, somebody will confirm it and you can then report it as such. BTW, time to upgrade...Go Wildcats! HTH, Marc Schwartz [OFFLIST and PRIVATE] Brad, A couple of comments. First, welcome to R. I hope that you enjoy it and find it of value. If you are not used to open source software and communities (ie. Linux, etc.), you will find that this community, unlike commercial paid support forums, tends to be direct with respect to comments. Don't take it personally. Be aware that nobody is getting paid to support R. It is developed and supported on a voluntary basis by a large body of folks, mainly those known as R Core. Some of them have quite literally risked their academic careers and livelihood to facilitate R's existence. You will, over time, get a flavor for the nature of the community and the interchange that takes place. As a result of the voluntary nature of the community, there is an a priori expectation that you will have put forth reasonable efforts to avail yourself of the various support resources before posting. Especially in the case of a bug report, as a member of R Core has to manually manage the handling and resolution of bug reports. A good place to start is to review the R Posting Guide: http://www.r-project.org/posting-guide.html which covers many of these issues and how to go about getting support via the various sources provided. That all being said, you will find that R's support mechanisms and resources are second to none and I would challenge any commercial software vendor to provide a comparable level of support and expertise. With respect to your specific question above and how the result is obtained: (a - b) / a [1] 0.0001019608 Here, 'a' is used as the scaling factor, since you only passed single values. If these were 'vectors' of values, the scaling factor would be impacted accordingly. As a result of R's open source nature, you have access to all of the source code that is R. You can download the source tarball (archive) from one of the CRAN mirrors, if you so desire. In this case, the actual function that is used is called all.equal.numeric(). This is a consequence of how R uses 'dispatch methods' after a call to a 'generic' function, such as all.equal(). If you are not familiar with these terms, the available R documentation is a good place to start, if you should decide to pursue moving into that level of detail. If you have experience in other programming languages, this may be second nature already. In many cases, R's functions are written in R itself. Others are written in FORTRAN and/or C that is compiled and linked to R via various calling mechanisms. Since R is an interpreted language, you can have easy access to many of the functions within the R console. Thus, at the R command prompt, you can type: all.equal.numeric [Note without the parens] which will then display a representation of the function's source code, enabling you to review how the function works. If you desire to become a better R user/programmer, this approach provides a reasonable way to see how functions are coded and to investigate algorithms and techniques. I hope that the above is helpful. Best regards, Marc __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Legend Title missing (PR#9226)
On Thu, 2006-09-14 at 20:35 +0200, [EMAIL PROTECTED] wrote: Full_Name: Mark O. Kimball Version: 2.1 OS: Linux Submission from: (NULL) (128.253.11.148) When log=x is used in a plot() function, the legend title option creates room for the text but does not display it. When the log=x function is disabled, the text is again displayed. The symbols and/or lines and legend text is not affected. From NEWS for R 2.3.1: o legend() with log axes would place the title in the wrong place. Upgrade to the current version. Please don't report bugs on obsolete versions. HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] graphics documentation omission (PR#9148)
On Tue, 2006-08-15 at 21:04 +0200, [EMAIL PROTECTED] wrote: ?axis says in Details: (for example, if the 'plot' argument 'asp' is set). The asp argument is not mentioned in ?plot and is not mentioned in ?par. It is referenced in ?plot.default and ?plot.window, with the former passing the 'asp' argument to the internal localWindow() function as a wrapper to the latter. In the HTML help page for ?axis, the 'asp' word in the quote above is a live link to ?plot.window. That is not helpful of course if one is working within ESS. Perhaps the See Also in ?axis should point to plot.default and plot.window? HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] graphics documentation omission (PR#9149)
On Tue, 2006-08-15 at 22:13 +0200, [EMAIL PROTECTED] wrote:
Thanks, Marc
Happy to help Rich.
Neither plot.window or plot.default are natural places for me to look for
argument names. par and plot are, for me, the natural places. I realize
that asp is not a par argument. Nonetheless, par is still the first place
to look for arguments that I would use in the ... position of any plot
function.
It would be easy to make a case that asp is the same type of argument as mar
or fin, thus it would make sense for it to be added to par.
Over the years, more from behavioral modification than from instinct,
I look at ?plot.default when I forget defacto plot() arguments. Of
course, plot.default() is in the See Also in ?plot.
It is rare for me to look at ?plot.window, given the rarity in which I
call it directly.
There are NO (that is, zero) live links in the Rgui help system.
Yeah. I enabled options(htmlhelp = TRUE) in my .Rprofile for the times
when I use R from the GNOME console, otherwise I am usually in ESS using
emacs 22 from CVS with the XFT patches.
I am not sure that I ever used the .CHM help when I used to run on
Windows, but presume that there is a hyperlink in that format to
parallel the live link in the HTML file.
help.search(asp) doesn't find anything related to graphics.
Could be added as a \concept presumably in the .Rd files for
plot.default and plot.window. For example:
\concept{asp aspect ratio}
ESS is much better for help than Rgui since ESS Rd mode can go to the help
page for any word that the cursor sits on. ESS for R help doesn't have the
links themselves.
Indeed.
Regards,
Marc
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Installation oddity on Fedora Core 5 (PR#8814)
On Thu, 2006-04-27 at 18:10 +0200, [EMAIL PROTECTED] wrote: Hello guys: I recently installed FC 5 linux and installed R from source. It installed fine, but there was an oddity that I want to report. Although I used ./configure --prefix=/usr/local/R-2.3.0 make it did not seem to recognize the prefix and went ahead and installed it in the source directory -- I usually ``tar xzf'' it to my home directory and run the installation scripts from there (as root) to install to /usr/local/R-2.3.0 This procedure worked fine on earlier versions of Fedora as well as on debian and other distros. Wonder if this is a gcc-4.1 or GNU make issue. Any feedback will be much appreciated. Tx. Sudipto. This question may be akin to Is the computer plugged in?, but did you also run: make install ? HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Installation oddity on Fedora Core 5 (PR#8814)
Hi Sudipto, OK. Just for clarification: 'make' will simply do the compilation 'in place' in the source directory tree. No installation is performed here. 'make install' will actually do the installation into the target folder. You can do 'make install' without 'make' first, but I suspect a lot of folks, myself included, will run 'make' first and then run 'make check-all' before installing, to be sure that the compilation is intact and passes all checks. In fact, I keep the output of 'make check-all' as part of my installation validation documentation. That being said, were there any error messages output during 'make install' that might suggest a problem such as access permission issues to the target directory or the like? You might also want to run (I think this is correct): dmesg in a console to look for any 'avc' error messages in the error log. This would be related to FC5's use of SELinux and there have been very significant SELinux related policy changes in FC5 that could feasibly be involved here also. HTH, Marc On Thu, 2006-04-27 at 11:25 -0500, Sudipto Banerjee wrote: Hi Marc: Thanks for your reply. Yeah, I did do make install and also tried it with simply make (which is, I presume, enough to produce the executable scripts for R). Please keep me posted. Best, S. On Thu, 27 Apr 2006, Marc Schwartz (via MN) wrote: On Thu, 2006-04-27 at 18:10 +0200, [EMAIL PROTECTED] wrote: Hello guys: I recently installed FC 5 linux and installed R from source. It installed fine, but there was an oddity that I want to report. Although I used ./configure --prefix=/usr/local/R-2.3.0 make it did not seem to recognize the prefix and went ahead and installed it in the source directory -- I usually ``tar xzf'' it to my home directory and run the installation scripts from there (as root) to install to /usr/local/R-2.3.0 This procedure worked fine on earlier versions of Fedora as well as on debian and other distros. Wonder if this is a gcc-4.1 or GNU make issue. Any feedback will be much appreciated. Tx. Sudipto. This question may be akin to Is the computer plugged in?, but did you also run: make install ? HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] sub returns garbage (PR#8687)
On Thu, 2006-03-16 at 22:28 +0100, [EMAIL PROTECTED] wrote:
Full_Name: Todd Bailey
Version: 2.1
OS: Mac OS-X 10.4.3
Submission from: (NULL) (87.112.79.124)
sub returns garbage in some strings when replacing something with nothing and
fixed=TRUE. For example:
a=c('hello','hello'); sub('lo','',a,fixed=TRUE)
[1] hel hel\0\0
a=c('hello','hello'); sub('lo','',a,fixed=FALSE)
[1] hel hel
This has been fixed in R Patched. From NEWS:
o sub(fixed = TRUE) could get wrong the length of the character
string of elements of the result after the first.
From my installation:
a - c('hello', 'hello'); sub('lo', '', a, fixed=TRUE)
[1] hel hel
a - c('hello', 'hello'); sub('lo', '', a, fixed=FALSE)
[1] hel hel
HTH,
Marc Schwartz
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] (Debian Bug 344248): R segfaults when pressing Delete (PR#8421)
On Wed, 2005-12-21 at 15:00 +0100, [EMAIL PROTECTED] wrote: --CE+1k2dSO48ffgeK Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Resend after type in mailheader. I have the bug on my system with yesterday's R 2.2.1. A library mismatch is still a likely cause. Dirk SNIP On 21 December 2005 at 01:50, Ethan Glasser-Camp wrote: | Package: r-base-core | Version: 2.2.1-1 | | When I run R, I can cause a crash by simply pressing Delete. Backspace works | fine, and I can use ESS without problems. Uh-oh. | [EMAIL PROTECTED]:~$ R | | R : Copyright 2005, The R Foundation for Statistical Computing | Version 2.2.1 (2005-12-20 r36812) | ISBN 3-900051-07-0 | | R is free software and comes with ABSOLUTELY NO WARRANTY. | You are welcome to redistribute it under certain conditions. | Type 'license()' or 'licence()' for distribution details. | | R is a collaborative project with many contributors. | Type 'contributors()' for more information and | 'citation()' on how to cite R or R packages in publications. | | Type 'demo()' for some demos, 'help()' for on-line help, or | 'help.start()' for an HTML browser interface to help. | Type 'q()' to quit R. | | [Previously saved workspace restored] Could you (temporarily) remove that file? | Segmentation fault | [EMAIL PROTECTED]:~$ That said, I just tried this where I seemed to have no ~/.RData file. Pressing DeletetReturn leads to a SegFault. I will pass this on R Core. Thanks for the bug report. Dirk | | | Here is a list of packages I thought might be relevant: | | $ dpkg -l r-* libreadline* | Desired=Unknown/Install/Remove/Purge/Hold | | Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed | |/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad) | ||/ Name VersionDescription | +++-==-==- | un libreadline-co none (no description available) | un libreadline-de none (no description available) | ii libreadline-ru 1.8.3+1.8.4pre Readline interface for Ruby 1.8 | ii libreadline4 4.3-18 GNU readline and history libraries, run-time | un libreadline4-d none (no description available) | ii libreadline5 5.1-1 GNU readline and history libraries, run-time | ii libreadline5-d 5.1-1 GNU readline and history libraries, developm | ii r-base 2.2.1-1GNU R statistical computing language and env | ii r-base-core2.2.1-1GNU R core of statistical computing language | un r-base-dev none (no description available) | ii r-base-html2.2.1-1GNU R html docs for statistical computing sy | ii r-base-latex 2.2.1-1GNU R LaTeX docs for statistical computing s | ii r-cran-boot1.2.24-1 GNU R package for bootstrapping functions fr | ii r-cran-cluster 1.10.2-1 GNU R package for cluster analysis by Rousse | ii r-cran-foreign 0.8.12-1 GNU R package to read/write data from other | ii r-cran-gdata 2.1.2-1GNU R package with data manipulation tools b | ii r-cran-gplots 2.2.0-1GNU R package with tools for plotting data b | un r-cran-gregmis none (no description available) | ii r-cran-gtools 2.2.2-1GNU R package with R programming tools by Gr | ii r-cran-kernsmo 2.22.15-1 GNU R package for kernel smoothing and densi | ii r-cran-lattice 0.12-11.1-1GNU R package for 'Trellis' graphics | ii r-cran-lattice 0.1.1-1GNU R package of additional graphical displa | ii r-cran-mgcv1.3-12-1 GNU R package for multiple parameter smoothi | ii r-cran-misc3d 0.3-1-1GNU R collection of 3d plot functions and rg | ii r-cran-nlme3.1.62-1 GNU R package for (non-)linear mixed effects | ii r-cran-quadpro 1.4.7-2GNU R package for solving quadratic programm | ii r-cran-rgl 0.65-1 GNU R package for three-dimensional visualis | ii r-cran-rpart 3.1.27-1 GNU R package for recursive partitioning and | ii r-cran-surviva 2.20-1 GNU R package for survival analysis | un r-cran-tkrplot none (no description available) | ii r-cran-vr 7.2.23-1 GNU R package accompanying the Venables and | un r-doc-html none (no description available) | ii r-doc-info 2.2.1-1GNU R info manuals statistical computing sys | un r-doc-pdf none (no description available) | un r-mathlib none (no description available) | ii r-recommended 2.2.1-1GNU R collection of recommended packages [me | | | Please let me know if there is anything else I can do to help. | | Ethan FWIW, I cannot replicate the segfault on FC4, with: Version 2.2.1 (2005-12-20 r36812) compiled from source yesterday. I have: $
Re: [Rd] package unlisted (PR#8410)
On Sun, 2005-12-18 at 15:31 +0100, [EMAIL PROTECTED] wrote: Full_Name: Cesar Henrique Torres Version: 2.2.0 OS: winXP-pro Submission from: (NULL) (201.1.196.50) Hi developer's! I'd like to know were gtools package is, because in the link below it isn't: http://cran.r-project.org/bin/windows/contrib/r-release/ This package is used in gplots and I'm trying to make a balloonplot. Thanks a lot, Cesar Torres. Please do not post bugs on Contributed packages to the main R-bugs repo. This now requires intervention by a member of R Core, none of whom are in a position to resolve the problem. You should contact the package maintainer instead. In this case it is Nitin Jain ([EMAIL PROTECTED]) who I have cc'd here. The problem here is that the current version of gtools for Windows is not passing the required checks before it can be allowed to be included in the main CRAN repo. This can be seen here: http://cran.r-project.org/bin/windows/contrib/checkSummaryWin.html On the above, scroll down to gtools and you will see links for the following error log and interim resolution: http://cran.r-project.org/bin/windows/contrib/2.2/check/gtools-check.log http://cran.r-project.org/bin/windows/contrib/2.2/last/ReadMe This brings you to: http://cran.r-project.org/bin/windows/contrib/2.2/last/ where you will find the last version of gtools for Windows that passed the required checks. This will need to be manually downloaded and may or may not work in the presence of the other related packages given the potential for version related conflicts. I suspect Nitin will have more definitive comments. HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Rgnome depends on obsolete components libglade/libxml (PR#8248)
On Mon, 2005-10-24 at 17:14 +0100, Hin-Tak Leung wrote: Peter Dalgaard wrote: [EMAIL PROTECTED] writes: Full_Name: Hin-Tak Leung Version: R 2.2.0 OS: x86 linux Submission from: (NULL) (131.111.126.242) Rgnome depends libglade 0.x and libxml 1.x . They are no longer shipped with Redhat EL4. (I know they are still in fedora core 4). Just a wish-list. But what is the wish? You mean get it upgraded to xml2 and glade2? Patches would likely be accepted... :-). libglade went from 0.17 to 1.99 then 2.x, but the current Rgnome code depends on some of 0.17-specific stuff. Already tried against libglade 2 - not trivial. It probably also requires changing to gtk2 also... Just like to have it in the distant TODO list, and not forgotten... Hin-Tak Leung According to the R Admin manual (2.2.0) on page 34: This interface is experimental and incomplete. The console offers a basic command line editing and history mechanism, along with tool and button bars that give a point-and-click console to some R commands. Many of the features of the console are currently stubs, and the console is **no longer under development**: it has been kept available as an example of adding a front-end to R. This language (my emphasis added) would suggest that a TODO list does not (or should not) exist...so Peter's suggestion would seem spot on. HTH, Marc Schwartz __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Trouble with ifelse and if statement (PR#7962)
On Wed, 2005-06-22 at 04:46 +0200, [EMAIL PROTECTED] wrote:
Full_Name: Woolton Lee
Version: 2.1
OS: windows
Submission from: (NULL) (128.118.224.46)
I did the following ('g' and 'h' are both numeric vectors)
i - abs(g-h)
creating a vector 'i' with values,
i
[1] 0.08 0.00 0.33 0.00 0.00 0.00 0.00 0.33 0.00 0.00 0.08 0.08 0.20 0.00
0.13
Now, I want to create a new vector =1 whenever 'i' = 0.33 and =0 otherwise. I
use the 'ifelse' function to do this but when I do, I find
ifelse(i==0.33,1,0)
[1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
even though the third and eighth elements of 'i' clearly equal 0.33. Unless I
have missed something, 'ifelse' should return a vector that looks like,
[1] 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0
When I try 0.13 I get a correct result
ifelse(i==0.13,1,0)
[1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
but when I try 0.08 or 0.2 I find a similar problem,
ifelse(i==0.08,1,0)
[1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ifelse(i==0.20,1,0)
[1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ifelse(i==0.2,1,0)
[1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Can you explain what is happening? Why is R incorretly returning a vector
that
does not reflect the values in 'i'? Is this due to a malfunction in R or
have I
missed something?
Thank you for your help.
Woolton
This is not a bug and yes you have missed something.
Read R FAQ 7.31 Why doesn't R think these numbers are equal?
More information is also available here:
http://grouper.ieee.org/groups/754/
One possible solution:
i
[1] 0.08 0.00 0.33 0.00 0.00 0.00 0.00 0.33 0.00 0.00 0.08 0.08 0.20
[14] 0.00 0.13
ifelse(sapply(i, function(x) all.equal(x, 0.33)) == TRUE, 1, 0)
[1] 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0
ifelse(sapply(i, function(x) all.equal(x, 0.08)) == TRUE, 1, 0)
[1] 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0
ifelse(sapply(i, function(x) all.equal(x, 0.2)) == TRUE, 1, 0)
[1] 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
Marc Schwartz
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
