An embedded and charset-unspecified text was scrubbed... Name: not available Url: https://stat.ethz.ch/pipermail/r-help/attachments/20050802/1aef7f8c/attachment.pl
Rgdal windows binary warning message
5 messages · Tony Gill, Lisa, Sean O'Riordain +2 more
Hello every one-- My sincere apologies if I have missed something obvious, but I need to construct some multinomial logit models. I remember using nnet for this in the past, but I am having trouble locating a OS X version of this (it's been a year or so since I had to use it). I found a Windows version, but when I try to install across platforms, it crashes R immediately--no error messages or anything helpful. Can any one direct me to a repository, or alternatively, to another package readily available Mac side that can do simple multinomial logit models? This is basic survey data--nothing fancy needed. All best, Lisa
Hi Lisa, as far as I know, nnet is part of the VR bundle which is a recommended package... refer http://www.stats.bris.ac.uk/R/src/contrib/Descriptions/VR.html so just try library(nnet) and see what happens! cheers! Sean
On 02/08/05, Lisa Schweitzer <lschwei at mac.com> wrote:
Hello every one-- My sincere apologies if I have missed something obvious, but I need to construct some multinomial logit models. I remember using nnet for this in the past, but I am having trouble locating a OS X version of this (it's been a year or so since I had to use it). I found a Windows version, but when I try to install across platforms, it crashes R immediately--no error messages or anything helpful. Can any one direct me to a repository, or alternatively, to another package readily available Mac side that can do simple multinomial logit models? This is basic survey data--nothing fancy needed. All best, Lisa
______________________________________________ R-help at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
This is a consequence of how VC++ compiled gdal. R is protecting itself against that. Since R did not crash, there is nothing to worry about.
On Tue, 2 Aug 2005, Tony Gill wrote:
Hi all, I just downloaded windows binaries of RGDAL (from http://www.stats.ox.ac.uk/pub/RWin/bin/windows/contrib/2.1/) and installed. I ran the example R_HOME\library\rgdal\R-ex\getPixmapGDAL.R and everything seemed to work as expected. However I got the following warning message: DLL attempted to change FPU control word from 8001f to 9001f The R-help (?dyn.load) describes the problem as: ************************************ External code must not change the floating point control word, but many DLLs do so. Common changes are to set it to use 53 bit precision instead of R's default 64 bit precision, or to unmask some exceptions. 'dyn.load' detects such changes, and restores R's control word to its default value of hex 8001F. This may cause the DLL to malfunction; if so, it should be rewritten to save and restore the control word itself. If 'warn.FPU' is set to 'TRUE' using the 'options' function, a warning will be printed. (If the warning says that the control word was changed from some other value than 8001F, please report the circumstances to the Windows maintainers: that probably indicates an internal bug.) ************************************ Does anyone know if this is a major problem for reading and manipulating image files using RDAL? And if so are there windows binaries which have patched it up? Thanks in advance Tony ****************************************** Tony Gill - PhD Candidate Centre for Remote Sensing & Spatial Information Science School of Geography, Planning & Architecture University of Queensland Brisbane, Queensland, AUSTRALIA, 4072 Ph: 61-7-3365-7027 email: t.gill1 at uq.edu.au [[alternative HTML version deleted]]
______________________________________________ R-help at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595
Prof Brian Ripley wrote:
This is a consequence of how VC++ compiled gdal. R is protecting itself against that. Since R did not crash, there is nothing to worry about.
To add to that: the RGDAL maintainer could work around this bug in VC++. I haven't used that compiler, but with others it is sufficient to ask R to set the FPU control word. Instructions are here: <http://www.stats.uwo.ca/faculty/murdoch/software/compilingDLLs/index.html#fpu> Duncan Murdoch
On Tue, 2 Aug 2005, Tony Gill wrote:
Hi all, I just downloaded windows binaries of RGDAL (from http://www.stats.ox.ac.uk/pub/RWin/bin/windows/contrib/2.1/) and installed. I ran the example R_HOME\library\rgdal\R-ex\getPixmapGDAL.R and everything seemed to work as expected. However I got the following warning message: DLL attempted to change FPU control word from 8001f to 9001f The R-help (?dyn.load) describes the problem as: ************************************ External code must not change the floating point control word, but many DLLs do so. Common changes are to set it to use 53 bit precision instead of R's default 64 bit precision, or to unmask some exceptions. 'dyn.load' detects such changes, and restores R's control word to its default value of hex 8001F. This may cause the DLL to malfunction; if so, it should be rewritten to save and restore the control word itself. If 'warn.FPU' is set to 'TRUE' using the 'options' function, a warning will be printed. (If the warning says that the control word was changed from some other value than 8001F, please report the circumstances to the Windows maintainers: that probably indicates an internal bug.) ************************************ Does anyone know if this is a major problem for reading and manipulating image files using RDAL? And if so are there windows binaries which have patched it up? Thanks in advance Tony ****************************************** Tony Gill - PhD Candidate Centre for Remote Sensing & Spatial Information Science School of Geography, Planning & Architecture University of Queensland Brisbane, Queensland, AUSTRALIA, 4072 Ph: 61-7-3365-7027 email: t.gill1 at uq.edu.au [[alternative HTML version deleted]]
______________________________________________ R-help at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html