Re: [R] fancyRpartPlot and the title at the bottom of the plot....
Hopefully, better late than never Current version of Rattle (3.1.4) supports sub= in fancyRpartPlot() to override or get rid of that useless title, as in: fancyRpartPlot(m, sub=) Install it using: install.packages(rattle, repos=http://rattle.togaware.com;, type=source) Graham Williams http://togaware.com On 6 January 2014 21:11, Levent TERLEMEZ lterle...@anadolu.edu.tr wrote: Dear Users, Is there way to avoid the useless title (Rattle, date-time and user imformation) at the bottom of the fancyRpartPlot. We already referencing it why should it be there? Thanks, Levent. [[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] Decision Tree
It is box.col= in prp(). fancyRpartPlot() in Rattle currently uses a fixed colour palette (something I should change). You could change it in a local copy of the fancyRpartPlot() code (see the line defining pals in the function): fancyRpartPlot I also give an example around page 33 of the Decision Trees chapter ( http://onepager.togaware.com/DTreesO.pdf) from the OnePageR website ( http://onepager.togaware.com). Graham Williams http://togaware.com On 11 July 2014 14:52, Abhinaba Roy abhinabaro...@gmail.com wrote: Hi Jean, I'd looked at the help for 'prp' but couldn't find the argument for changing box colours. Am I missing something? On Thu, Jul 10, 2014 at 8:01 PM, Adams, Jean jvad...@usgs.gov wrote: The function fancyRpartPlot() is actually in the rattle package, and it is a wrapper for the prp() function in the rpart.plot package. If you look at the help for prp(), you should be able to see how to change the color. library(rpart.plot) ?prp Jean On Thu, Jul 10, 2014 at 12:34 AM, Abhinaba Roy abhinabaro...@gmail.com wrote: Hi R-helpers, Is it possible to change the color of the boxes when plotting decision trees using 'fancyRpartPlot()' from rpart.plot package ? -- Regards, Abhinaba Roy [[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. -- Regards, Abhinaba Roy [[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] Basket Analysis in R: extract rules
Try: inspect(sort(crs$apriori, by=support)) Graham Williams http://togaware.com On 27 June 2014 16:46, Abhinaba Roy abhinabaro...@gmail.com wrote: Dear R-helpers, I have run a basket analysis in Rattle. I've used 'arules' package. crs$apriori - apriori(crs$transactions, parameter = list(support=0.100, confidence=0.100, minlen=2)) and str(crs$apriori) Formal class 'rules' [package arules] with 4 slots ..@ lhs:Formal class 'itemMatrix' [package arules] with 3 slots .. .. ..@ data :Formal class 'ngCMatrix' [package Matrix] with 5 slots .. .. .. .. ..@ i : int [1:22] 0 3 0 1 2 3 2 1 3 1 ... .. .. .. .. ..@ p : int [1:17] 0 1 2 3 4 5 6 7 8 9 ... .. .. .. .. ..@ Dim : int [1:2] 4 16 .. .. .. .. ..@ Dimnames:List of 2 .. .. .. .. .. ..$ : NULL .. .. .. .. .. ..$ : NULL .. .. .. .. ..@ factors : list() .. .. ..@ itemInfo :'data.frame': 4 obs. of 1 variable: .. .. .. ..$ labels:Class 'AsIs' chr [1:4] 1 2 3 4 .. .. ..@ itemsetInfo:'data.frame': 0 obs. of 0 variables ..@ rhs:Formal class 'itemMatrix' [package arules] with 3 slots .. .. ..@ data :Formal class 'ngCMatrix' [package Matrix] with 5 slots .. .. .. .. ..@ i : int [1:16] 3 0 1 0 3 2 1 2 1 3 ... .. .. .. .. ..@ p : int [1:17] 0 1 2 3 4 5 6 7 8 9 ... .. .. .. .. ..@ Dim : int [1:2] 4 16 .. .. .. .. ..@ Dimnames:List of 2 .. .. .. .. .. ..$ : NULL .. .. .. .. .. ..$ : NULL .. .. .. .. ..@ factors : list() .. .. ..@ itemInfo :'data.frame': 4 obs. of 1 variable: .. .. .. ..$ labels:Class 'AsIs' chr [1:4] 1 2 3 4 .. .. ..@ itemsetInfo:'data.frame': 0 obs. of 0 variables ..@ quality:'data.frame': 16 obs. of 3 variables: .. ..$ support : num [1:16] 0.16 0.16 0.169 0.169 0.321 ... .. ..$ confidence: num [1:16] 0.619 0.244 0.654 0.242 0.661 ... .. ..$ lift : num [1:16] 0.943 0.943 0.935 0.935 1.007 ... ..@ info :List of 4 .. ..$ data : language crs$transactions .. ..$ ntransactions: int 894 .. ..$ support : num 0.1 .. ..$ confidence : num 0.1 How can I extract the set of rules from crs$apriori? -- Regards Abhinaba Roy [[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] Correlation in Rattle
Thanks for assisting here Paul. It turns out there is a bug in this release of Rattle. It is fixed and 2.6.19 will include the fix. In the mean time you can download an updated 2.6.18 from http://www.togaware.com.au/repository/ Regards, Graham On 19 May 2012 01:14, Paul Miller pjmiller...@yahoo.com wrote: Hi Avideh, Sorry to hear you're still having a problem. Tried usinging this two options on my 64-bit and 32-bit versions of R 2.15.0. Everything works fine. So I've at least been able to determine that there's nothing wrong with Rattle or with the combination of Ratttle and either the 64-bit and 32-bit versions of R 2.15.0. Beyond that, I'm not sure how much I can help you. Tried to think of silly things that could be happening. So, for example, if you don't click on Settings - Use Cairo Graphics Device, graphs generally get printed in R itself rather than on top of the Rattle graphical interface. If that were happening, then simply minimizing Rattle and looking at R would reveal the graph. I think this is unlikely though, given that you seem to be able to view graphs created using Distributions on the Explore tab. The only thing I can think of other than that is that maybe there is something wrong with your installation of R or with the process you're using to install Rattle. I don't recall there being anything tricky about installing either R or Rattle though. So I don't know what the problem could be. You could try contacting Graham Williams to see what he suggests. Or you could turn this over to an IT person if you have one. Sorry I can't be of more help. Paul --- On Fri, 5/18/12, avideh yesharim avidehyesha...@yahoo.com wrote: From: avideh yesharim avidehyesha...@yahoo.com Subject: Re: Correlation in Rattle To: Paul Miller pjmiller...@yahoo.com Received: Friday, May 18, 2012, 9:01 AM Hi Paul, Thank you so much for you response. I tried using the Weather data and still had the name problem. I also removed the package and reinstalled it and no luck with that either. All the other commands work, the only 2 that do not work are Correlation and Principal Components. When I select either of them and click Execute absolutely nothing happens, I don't even get a warning message. I am working on a new project and since I don't have a great knowledge of R programing I am highly dependent on Rattle and it really bothers me that these two commands are not working for me. I hope we could resolve this issue. Best Regards, Avideh From: Paul Miller pjmiller...@yahoo.com To: avideh yesharim avidehyesha...@yahoo.com Cc: r-help@r-project.org Sent: Friday, May 18, 2012 9:15 AM Subject: Re: Correlation in Rattle Hi Avideh, I have Rattle installed on the 64-bit version of R 2.15.0 and this seems to work OK. I had a dataset of my own loaded into Rattle when I saw your email. Tried creating the plot using these data but got a warning saying I should limit the number of variables to 40. So I changed to the weather dataset that comes with the software (which has fewer variables) and everything worked fine. Do you have a lot of variables in your data? What happens when you try this using the weather dataset? If you're unfamiliar with the weather dataset, you can load it by launching Rattle and then clicking execute. If there really is a problem, maybe uninstalling and reinstalling Rattle would fix it? For uninstall, you can use code like: remove.packages(rattle). For re-install, you could use code like: install.packages(rattle, dependencies = TRUE). HTH, Paul __ 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@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] pmml for random forest rules
Hi Patrick, Thanks for the detailed report. See comments below. On 11 October 2011 05:57, Patrick McCann patmmcc...@gmail.com wrote: [...] I am having some trouble using R 2.13.1 for generating a pmml object of class c('randomForest.formula', 'randomForest') [...] Random Forest (and randomSurvivalForest) — randomForest (Breiman and Cutler. R port by A. Liaw and M. Wiener, 2009) and randomSurvivalForest (Ishwaran and Kogalur , 2009): PMML export of a randomSurvivalForest rsf object. This function gives the user the ability to export PMML containing the geometry of a forest. [...] Error in UseMethod(pmml) : no applicable method for 'pmml' applied to an object of class c('randomForest.formula', 'randomForest') Sorry for the ambiguity there. It tries to say in the paper that pmml supports PMML export of a randomSurvivalForest rsf object. It mentions randomForest but does not say it can export randomForest. There is some experimental code for pmml.randomForest but it has not yet been completed. Also, if I run these lines of code data(Adult) ## Mine association rules. rules - apriori(Adult, parameter = list(supp = 0.5, conf = 0.9, target = rules)) pmml(rules) I get this error: pmml(rules) Error in function (classes, fdef, mtable) : unable to find an inherited method for function size, for signature itemMatrix [...] standardGeneric(size), environment) 3: size(is.unique) 2: pmml.rules(rules) 1: pmml(rules) That's odd. Not quite sure yet what is causing that. On my system it works just fine: library(pmml) library(arules) data(Adult) rules - apriori(Adult, parameter = list(supp = 0.5, conf = 0.9, target = rules)) pmml(rules) PMML version=3.2 ... Header copyright=Copyright (c) 2011 gjw... Extension name=user value=gjw extender=Rattle/PMML/ Application name=Rattle/PMML version=1.2.27/ Timestamp2011-10-11 21:50:40/Timestamp /Header [...] My system: rattleInfo() Rattle: version 2.6.11 cran 2.6.11 R: version 2.13.2 (2011-09-30) (Revision 57111) Sysname: Linux Release: 2.6.38-12-generic Version: #51-Ubuntu SMP Wed Sep 28 14:27:32 UTC 2011 [...] pmml: version 1.2.27 [...] arules: version 1.0-6 I'm using R 2.13.2 - could that be an issue - you have 2.13.1? Regards, Graham __ 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] Import/convert PMML to R model
Not possible (at least with the pmml package) at this time. There is some experimental code for reading PMML (and converting into standalone executable C code) but importing into an R object needs quite a bit of work to re-create the kmeans object before it would be worth releasing. Regards, Graham On 1 June 2011 18:40, Raji raji.sanka...@gmail.com wrote: Hi R-helpers, Can you please let me know if it is possible to import a PMML in R? If yes, can you give me the command to do the same? If not, can you tell me the reason why? Many thanks, Raji -- View this message in context: http://r.789695.n4.nabble.com/Import-convert-PMML-to-R-model-tp3332772p3565260.html Sent from the R help mailing list archive at Nabble.com. __ 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@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] Launcher for Rattle?
Thanks to the advice from Brian, the next version of Rattle (2.6.7) will have the two bugs fixed. Then your launcher will simply need to be: sh -c 'R_DEFAULT_PACKAGES=rattle R $@' Regards, Graham On 16 April 2011 20:57, Iurie Malai iurie.ma...@gmail.com wrote: Thank you! Your code worked! I tested all the combinations and this shows that, in addition to the 'rattle', in this launcher code the 'utils' package is required, otherwise 'rattle' do not launch. Without all other packages 'rattle' starts very well. But what will be better, the inclusion or not of the 'datasets', 'grDevices', 'graphics', 'stats' and 'methods' packages in this launcher code? I do not understand if they are needed for proper functioning of the 'rattle'. Regards, Iurie 2011/4/16 Prof Brian Ripley This is a result of bugs in rattle. The first bug is that you are explicitly asked on its help page not to use installed.packages() in that way. Second, packages should be able to be loaded in a session with just base loaded, so rattle's startup code should have utils::installed.packages(). (In this case it is invoked from rattle(), AFAICS, hence not invoked in the loading tests in R CMD check.) However, your startup code is wrong: from ?options defaultPackages: the packages that are attached by default when R starts up. Initially set from value of the environment variable R_DEFAULT_PACKAGES, or if that is unset to c(datasets, utils, grDevices, graphics, stats, methods). (Set R_DEFAULT_PACKAGES to NULL or a comma-separated list of package names.) ^^^ And by default, R_DEFAULT_PACKAGES is unset. Try sh -c 'R_DEFAULT_PACKAGES=datasets,utils,grDevices,graphics,stats,rattle R $@' (You may or may not need 'methods' as well.) [[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] No response after click the show Rules button on Tab Associate.
On 10 March 2011 02:07, Xiaobo Gu guxiaobo1...@gmail.com wrote: set transactions ...[35 item(s), 8 transaction(s)] done [0.00s]. That does not look right? I think it's because there are to few sample records, so all the rules are with 100% confidence Sorry - I think you might have misunderstood - have a look at the data - there are not 35 items - there are 8 items. There are 35 transactions. You have the wrong variables selected. You need to . Be sure to make items your Target and tx_no your Ident rather than the other way around. Regards, Graham __ 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] No response after click the show Rules button on Tab Associate.
On 12 March 2011 00:07, Xiaobo Gu guxiaobo1...@gmail.com wrote: On Fri, Mar 11, 2011 at 2:55 AM, Graham Williams graham.willi...@togaware.com wrote: Did you scroll down the window to see the rules? OK, it takes a long time for rattle to show the rules, about 30 seconds, and why the message on the status bar is the decision tree model has been built. Time taken:0.01 secs, I have attached the little dataset used to do the test. I've fixed the incorrect decision tree text. Thanks. For the other issue see below. And thanks for supplying a repeatable sample and dataset. Another question, why the confidence of the rules are all 1 ? On 10 March 2011 02:07, Xiaobo Gu guxiaobo1...@gmail.com wrote: set transactions ...[35 item(s), 8 transaction(s)] done [0.00s]. That does not look right? Be sure to make items your Target and tx_no your Ident rather than the other way around. Regards, Graham __ 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] No response after click the show Rules button on Tab Associate.
Did you scroll down the window to see the rules? Regards, Graham On 10 March 2011 02:07, Xiaobo Gu guxiaobo1...@gmail.com wrote: Hi, I am using Rattle 2.6.4 with R 2.12.2 on win64, is this a bug ? Following is the content after execute the associate analysis process: Summary of the Apriori Association Rules: Number of Rules: 23351 Summary of the Measures of Interestingness: support confidence lift Min. :0.1250 Min. :1 Min. :2.667 1st Qu.:0.1250 1st Qu.:1 1st Qu.:2.667 Median :0.1250 Median :1 Median :4.000 Mean :0.1314 Mean :1 Mean :3.983 3rd Qu.:0.1250 3rd Qu.:1 3rd Qu.:4.000 Max. :0.3750 Max. :1 Max. :8.000 Summary of the Execution of the Apriori Command: parameter specification: confidence minval smax arem aval originalSupport support minlen maxlen target ext 0.8 0.1 1 none FALSE TRUE 0.1 1 10 rules FALSE algorithmic control: filter tree heap memopt load sort verbose 0.1 TRUE TRUE FALSE TRUE 2 TRUE apriori - find association rules with the apriori algorithm version 4.21 (2004.05.09) (c) 1996-2004 Christian Borgelt set item appearances ...[0 item(s)] done [0.00s]. set transactions ...[35 item(s), 8 transaction(s)] done [0.00s]. sorting and recoding items ... [35 item(s)] done [0.00s]. creating transaction tree ... done [0.00s]. checking subsets of size 1 2 3 4 5 6 7 8 9 10 done [0.00s]. writing ... [23351 rule(s)] done [0.00s]. creating S4 object ... done [0.01s]. Time taken: 0.01 secs Rattle timestamp: 2011-03-09 23:00:14 dell == Xiaobo Gu __ 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] problem installing R in Ubuntu 10.04 -HELP
I have seen this issue reported by others though I don't know what the root cause is. Others have solved it by removing the .Rdata file causing the problem (there must be one if it says so). It would usually, I think, be in the folder where you start R. Regards, Graham On 20 February 2011 13:19, fdp fdp2...@gmail.com wrote: I am running Ubuntu 10.04 (on a Windows machine with dual boot) and I having a very hard time trying to recover from what I initially thought was a minor problem. I was trying to install rattle() and it failed, and after that I cannot get R to run AT ALL! I've tried multiple times to reinstall it with a clean apt-get removal and install and nothing... I keep get the following message: *Error in loadNamespace(name) : there is no package called 'rattle' Fatal error: unable to restore saved data in .RData * I've tried to remove all .RData files, but couldn't find many... I even remove all R files/directories/config-files and still nothing worked. Has anyone encounter something like this? ANYTHING HELPS!!! Thanks a lot... fdp [[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@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] Problem in installing and starting Rattle
On 15 November 2010 15:30, kgorahava kgorah...@gmail.com wrote: Hello, I am trying to install and run Rattle on my Dell Laptop and I have Windows 7 OS. The following three commands executed successfully : install.packages(RGtk2) install.packages(rattle) library(rattle) Rattle: Graphical interface for data mining using R. Version 2.5.47 Copyright (c) 2006-2010 Togaware Pty Ltd. Type 'rattle()' to shake, rattle, and roll your data. When I type the below command, I get an error message. rattle() Error in inDL(x, as.logical(local), as.logical(now), ...) : unable to load shared object 'C:/Users/Kaushik/Documents/R/win-library/2.12/RGtk2/libs/i386/RGtk2.dll': LoadLibrary failure: The specified module could not be found. Failed to load RGtk2 dynamic library, attempting to install it. trying URL ' http://downloads.sourceforge.net/gtk-win/gtk2-runtime-2.22.0-2010-10-21-ash.exe?download ' Content type 'application/octet-stream' length 7820679 bytes (7.5 Mb) opened URL downloaded 7.5 Mb Learn more about GTK+ at http://www.gtk.org If the package still does not load, please ensure that GTK+ is installed and that it is on your PATH environment variable IN ANY CASE, RESTART R BEFORE TRYING TO LOAD THE PACKAGE AGAIN Error in as.GType(type) : Cannot convert RGtkBuilder to GType rattle() Error in as.GType(type) : Cannot convert RGtkBuilder to GType Why do I get this error message ? Please advice. It looks like you did not install the gtk2 package (independent of R) first. But in the process, RGtk2 noticed this and it looks like it installed it okay. You might like to follow the instructions that are shouted at you as all capitals: IN ANY CASE, RESTART R BEFORE TRYING TO LOAD THE PACKAGE AGAIN Hope that works. Installation instructions for Rattle can be found at: http://datamining.togaware.com/survivor/Install_MS_Windows.html Regards, Graham [[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] Problem in installing and starting Rattle
On 16 November 2010 02:40, Feng Mai maif...@gmail.com wrote: I also have the problem trying to start rattle Windows 7 32-bit R 2.12.0 When I try library(rattle) I get an error message The procedure entry point deflateSetHeader could not be located in the dynamic link library zilb1.dll I hit OK and it prompts me to install GTK+ again. I tried to uninstall GTK+ first and delete all related files as Dieter suggested but still wont work :( Yes, this is another issue that arose with R 2.12.0 and discussed on the rattle-users mailing list (http://groups.google.com/group/rattle-users). Be sure to use the newest version of Rattle (currently 2.5.47) where this has been fixed. I've not worked out why, but if the XML package is loaded before the RGtk2 package we see this error, but not if RGtk2 is loaded first. Rattle now loads RGtk2 first. Regards, Graham -- View this message in context: http://r.789695.n4.nabble.com/Problem-in-installing-and-starting-Rattle-tp3042502p3043262.html Sent from the R help mailing list archive at Nabble.com. __ 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] Rattle not reading files on windows 7
IGNACIO PEREZ VELEZ ignacio.perez at escuelaing.edu.co writes: Hi, I migrated to Windows 7 and now rattle does not read files whenever I try to read a file I get the following error: Error en sqrt(ncol(crs$dataset)) : Argumento no numérico para función matemática In english: Error on sqrt(ncol(crs$dataset)) : Non numeric argument for a mathematical function Any clue on what is going on? Thanks, Ignacio Hi Ignacio, Some more details would be useful. Are you trying to load a CSV file? There were some recent changes (added support for .xls and .xlxs files) that could be the culprit here. We can take the debugging off-line if you wanted to respond directly to me. Regards, Graham __ 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] rattle(): unable to load shared library
Sometimes I've seen this under MS/Windows and the solution has been to reinstall the gtk libraries independent of R and to make sure this is the one installed: http://downloads.sourceforge.net/gladewin32/gtk-2.12.9-win32-2.exe. It looks like glade was not in the gtk libraries you installed? Hope that helps. Regards, Graham 2009/11/26 Matthieu Stigler matthieu.stig...@gmail.com Hi I'm trying to install the rattle GUI on winwos 2000, with last version R 2.10. I had a first problem using the rattle package, as it was asking pkg XMl, which is no more in the CRAN repo. I instead simply install in R/libraries the XML pkg from: http://www.stats.ox.ac.uk/pub/RWin/bin/windows/contrib/2.10/ Maybe I mised something here... So now I was able to run the library(rattle) then, using: rattle() I have following error message, it wants to install GTK+, what I do, restart R but then it does not work and I have following error message: Error in inDL(x, as.logical(local), as.logical(now), ...) : unable to load shared library 'C:/Programme/R/R-2.10.0/library/RGtk2/libs/RGtk2.dll': LoadLibrary failure: Die angegebene Prozedur wurde nicht gefunden. Failed to load RGtk2 dynamic library, attempting to install it. Learn more about GTK+ at http://www.gtk.org If the package still does not load, please ensure that GTK+ is installed and that it is on your PATH environment variable IN ANY CASE, RESTART R BEFORE TRYING TO LOAD THE PACKAGE AGAIN Error in .Call(name, ..., PACKAGE = PACKAGE) : C symbol name S_glade_xml_new not in DLL for package RGtk2 Rattle timestamp (for the error above): 2009-11-25 17:47:50 Checked for the install of GTK+, it's on the disk, and even on the PATH... don't know what I should do... any idea? Thanks!! Matthieu Stigler [[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] Has anyone had success with RGTK2/rattle and windows 7 64-bit?
I don't have a Windows 7 to test this on yet - works on Vista and XP. Did you install the GTK libraries (separately to R)? Regards, Graham 2009/11/21 Tetrick, Scott scott.tetr...@intel.com I have been unable to get rattle to run in my new Windows7-64 bit configuration. For wither Rgtk2 or rattle, I get an error: Entry point not found The procedure entry point g_assetion_message_expr could not be located in the dynamic link library libglib-2.0-0.dll. Any help is really appreciated. __ 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] RandomForest PMML
2009/11/4 NCS nc1...@yahoo.com: I cannot seem to write a randomforest model in PMML - either through calling PMML(model) or through Rattle. It appears that it is not yet supported. Randomsurvivalforest is, but not randomforest. Any ideas on possible workarounds for this? Thanks ncs It is not yet implemented. I am waiting on some time to become available to do so. Regards, Graham __ 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] Regarding Performance and Prediction routines of rattle library and XML package
Try the ROCR package. Regards, Graham 2009/9/13 Abbas R. Ali abba...@yahoo.com Hi Can anybody tell me in which library Performance and Prediction routines exist to find AUC and I am unable to find a dependency of rattle library, XML, for Windows can any body tell me about that. Thanks [[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] Problem with RGtk2 Rattle
The tooltip error is a known issue - old versions of GTK (or perhaps RGtk2) did not have the function I am now using to install tooltips. You need to follow Felix's advice and install a new version of GTK (nothing to do with R), then install R - the order used to be important. I would suggest starting over again - unistall everything and then follow the instructions precisely. I don't think the SVG is an issue. As a temporary fix, try: library(rattle) crv$load.tooltips - FALSE rattle() Regards, Graham 2009/8/2 Wayne Murray wayne.mur...@medicareaustralia.gov.au hI aLL I feel with the great advice I am getting from Felix Graham, that I am finally making progress I feel that finally I have identified the REAL PROBLEM. Of cause I am unsure how I will solve it but that is another story. The trace follows after following Grahams latest advice Error in normalizePath(path) : path[1]=C:\Program Files\R\R-2.9.1\library/RSvgDevice: The system cannot find the file specified local({pkg - select.list(sort(.packages(all.available = TRUE))) + if(nchar(pkg)) library(pkg, character.only=TRUE)}) Loading required package: pmml Loading required package: XML Rattle: Graphical interface for data mining using R. Version 2.4.86 Copyright (C) 2006-2009 Togaware Pty Ltd. Type rattle() to shake, rattle, and roll your data. rattle() Error in .RGtkCall(R_setGObjectProps, obj, value, PACKAGE = RGtk2) : Invalid property tooltip-text! There appears to be a problem with the path for the RSvgDevice, but I don't know where that is stored how I can change / edit the path Thanks for your patience and understanding Wayne Graham Williams wrote: 2009/8/1 Felix Andrews fe...@nfrac.org Um, it sounds as if you are trying to install glade-3.4.3-win32.zip into R as a package... but it is not an R package!! GTK+/Glade is a system library to be installed into Windows. You should download the .exe (not the .zip) http://downloads.sourceforge.net/gladewin32/gtk-dev-2.12.9-win32-2.exe and run it to install it. By the way, problems with rattle are best sent to the rattle-users mailing list: http://groups.google.com/group/rattle-users -Felix Also Wayne, I hope you are following the instructions at http://datamining.togaware.com/survivor/Install_MS_Windows.html If there is any ambiguity there please let me know so I can make it clearer. I know of many who have installed Rattle following these, so they should work. (And thanks for the help Felix.) Regards, Graham -- Felix Andrews / å®â°Ã§Â¦ ç«⹠Postdoctoral Fellow Integrated Catchment Assessment and Management (iCAM) Centre Fenner School of Environment and Society [Bldg 48a] The Australian National University Canberra ACT 0200 Australia M: +61 410 400 963 T: + 61 2 6125 1670 E: felix.andr...@anu.edu.au CRICOS Provider No. 00120C -- http://www.neurofractal.org/felix/ __ 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. - Dr D. W. Murray Canberra, Australia -- View this message in context: http://www.nabble.com/Problem-with-RGtk2---Rattle-tp24734447p24775223.html Sent from the R help mailing list archive at Nabble.com. __ 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] Problem with RGtk2 Rattle
2009/8/1 Felix Andrews fe...@nfrac.org Um, it sounds as if you are trying to install glade-3.4.3-win32.zip into R as a package... but it is not an R package!! GTK+/Glade is a system library to be installed into Windows. You should download the .exe (not the .zip) http://downloads.sourceforge.net/gladewin32/gtk-dev-2.12.9-win32-2.exe and run it to install it. By the way, problems with rattle are best sent to the rattle-users mailing list: http://groups.google.com/group/rattle-users -Felix Also Wayne, I hope you are following the instructions at http://datamining.togaware.com/survivor/Install_MS_Windows.html If there is any ambiguity there please let me know so I can make it clearer. I know of many who have installed Rattle following these, so they should work. (And thanks for the help Felix.) Regards, Graham -- Felix Andrews / å®ç¦ç« Postdoctoral Fellow Integrated Catchment Assessment and Management (iCAM) Centre Fenner School of Environment and Society [Bldg 48a] The Australian National University Canberra ACT 0200 Australia M: +61 410 400 963 T: + 61 2 6125 1670 E: felix.andr...@anu.edu.au CRICOS Provider No. 00120C -- http://www.neurofractal.org/felix/ __ 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] Question about rpart decision trees (being used to predict customer churn)
2009/7/27 Robert Smith robertpsmith2...@gmail.com Hi, I am using rpart decision trees to analyze customer churn. I am finding that the decision trees created are not effective because they are not able to recognize factors that influence churn. I have created an example situation below. What do I need to do to for rpart to build a tree with the variable experience? My guess is that this would happen if rpart used the loss matrix while creating the tree. experience - as.factor(c(rep(good,90), rep(bad,10))) cancel - as.factor(c(rep(no,85), rep(yes,5), rep(no,5), rep(yes,5))) table(experience, cancel) cancel experience no yes bad 5 5 good 85 5 rpart(cancel ~ experience) n= 100 node), split, n, loss, yval, (yprob) * denotes terminal node 1) root 100 10 no (0.900 0.100) * I tried the following commands with no success. rpart(cancel ~ experience, control=rpart.control(cp=.0001)) rpart(cancel ~ experience, parms=list(split='information')) rpart(cancel ~ experience, parms=list(split='information'), control=rpart.control(cp=.0001)) rpart(cancel ~ experience, parms=list(loss=matrix(c(0,1,1,0), nrow=2, ncol=2))) Thanks a lot for your help. Best regards, Robert Hi Robert, Perhaps try a less extreme loss matrix: rpart(cancel ~ experience, parms=list(loss=matrix(c(0,5,1,0), byrow=TRUE, nrow=2))) Output from Rattle: Summary of the Tree model for Classification (built using rpart): n= 100 node), split, n, loss, yval, (yprob) * denotes terminal node 1) root 100 50 no (0.9000 0.1000) 2) experience=good 90 25 no (0.9444 0.0556) * 3) experience=bad 10 5 yes (0.5000 0.5000) * Classification tree: rpart(formula = cancel ~ ., data = crs$dataset, method = class, parms = list(loss = matrix(c(0, 5, 1, 0), byrow = TRUE, nrow = 2)), control = rpart.control(cp = 0.0001, usesurrogate = 0, maxsurrogate = 0)) Variables actually used in tree construction: [1] experience Root node error: 50/100 = 0.5 n= 100 CP nsplit rel error xerror xstd 1 0.4000 0 1.01.0 0.30 2 0.0001 1 0.60.6 0.22 TRAINING DATA Error Matrix - Counts Actual Predicted no yes no 85 5 yes 5 5 TRAINING DATA Error Matrix - Percentages Actual Predicted no yes no 85 5 yes 5 5 Time taken: 0.01 secs Generated by Rattle 2009-08-02 08:24:50 gjw == [[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] Problem with RGtk2 Rattle
Hi Wayne, I have seen reports of that error at times on MS/Windows. This beta version should have fixed it: install.packages(rattle, repos=http://rattle.togaware.com;) Let me know how you go and I'll upload the new version to CRAN. It would also be useful to know what version or Rattle and RGtk2 and Operating System you are using. Regards, Graham 2009/7/30 Wayne Murray wayne.mur...@medicareaustralia.gov.au HI Apologies for previously trying to post this question onto the Dev forum. I have recently update my versions of R and related packages. When I try to use rattle the following message appears Error in .RGtkCall(R_setGObjectProps, obj, value, PACKAGE = RGtk2) : Invalid property tooltip-text! I have downloaded and installed the latest available version of RGtk2, so I am at a loss to explain this error, or more importantly what I need to do to overcome it Thanks for any suggestions Regards Wayne - Dr D. W. Murray Canberra, Australia -- View this message in context: http://www.nabble.com/Problem-with-RGtk2---Rattle-tp24734447p24734447.html Sent from the R help mailing list archive at Nabble.com. __ 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] Rattle: Trouble opening .csv files
Hi James, I'm assuming this might be on MS/Windows? Could you try: install.packages(rattle, repos=http://rattle.togaware.com;) This beta may have fixed that problem. Let me know how you go. Regards, Graham 2009/6/30 Thompson, James james.thomp...@penfed.org I am a novice R user and recently installed Rattle primarily for the decision tree functionality. I immediately run into problems trying to open a .csv file from the Data tab. The error message I get in my R log window is: Error in if (crv$load.tooltips) loadTooltips() : argument is of length zero Help would be appreciated. Jim Thompson [[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] Problem with RGtk2 Rattle
Hi Wayne - but what version of the other tools have you installed? Regards, Graham 2009/7/30 Wayne Murray wayne.mur...@medicareaustralia.gov.au HI Graham Thanks for responding so promptly - unfortunately downloading and running this new version of Rattle did not alter the outcome - I am however running on Windows XP Regards Wayne Wayne Murray wrote: HI Apologies for previously trying to post this question onto the Dev forum. I have recently update my versions of R and related packages. When I try to use rattle the following message appears Error in .RGtkCall(R_setGObjectProps, obj, value, PACKAGE = RGtk2) : Invalid property tooltip-text! I have downloaded and installed the latest available version of RGtk2, so I am at a loss to explain this error, or more importantly what I need to do to overcome it Thanks for any suggestions Regards Wayne - Dr D. W. Murray Canberra, Australia -- View this message in context: http://www.nabble.com/Problem-with-RGtk2---Rattle-tp24734447p24736985.html Sent from the R help mailing list archive at Nabble.com. __ 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] RODBC crashes connecting to Teradata
Received Fri 19 Dec 2008 7:38am +1100 from Prof Brian Ripley: On Thu, 18 Dec 2008, Francisco Javier Perez Caballero wrote: Hi, I'm trying to connect to a Teradata database via RODBC on a Linux 64 machine (Red Hat Enterprise Linux 5). [...] library(RODBC) conn = odbcConnect(thedsn, uid=theuid, pwd=thepass) *** caught segfault *** address (nil), cause 'memory not mapped' [...] Most likely a broken driver. It seems that several do not support GetInfo correctly: isql does not use it by default. Debugging will show you where the error actually was. I came across the same problem recently and could not find any discussion on the r-sig-db list solving this. Here's my solution. I have a similar setup - Teradata ODBC driver on amd64 GNU/Linux (Debian), using the RODBC package, and getting a segfault on odbcConnect. odbcConnect calls odbcDriverConnect which in turn .Call's C_RODBCDriverConnect, which actually succeeds. Then there is a .Call to C_RODBCGetInfo which results in the segfault. I suspect underneath the Teradata driver is crashing on querying a table that does not exists or that you may not have access to (for me the default odbc GetInfo seems to try to access the table DBC.Table, but our setup does not have that available, with DBC.TableX being used instead!) I dropped the GetInfo call from odbcDriverConnect and all is well, until I query a table that does not exist, and once again the Teradata driver crashes (also crashes in isql when typing help or querying a table that does not exist). Regards, Graham In case it helps, I use this to connect. channel - odbcDriverConnect(DSN=mydsn;UID=user;PWD=password) with the following function. odbcDriverConnect - function (connection = , case = nochange, believeNRows = TRUE, colQuote, tabQuote = colQuote, DBMSencoding = , rows_at_time = 1000, bulk_add = NULL) { id - as.integer(1 + runif(1, 0, 1e+05)) stat - .Call(RODBC:::C_RODBCDriverConnect, as.character(connection), id, as.integer(believeNRows)) if (stat 0) { warning(ODBC connection failed) return(stat) } ### res - .Call(C_RODBCGetInfo, attr(stat, handle_ptr)) ### Also removed references to res in the following. if (missing(colQuote)) colQuote - \ if (missing(case)) case - nochange switch(case, toupper = case - 1, oracle = case - 1, tolower = case - 2, postgresql = case - 2, nochange = case - 0, msaccess = case - 0, mysql = case - ifelse(.Platform$OS.type == windows, 2, 0), stop(Invalid case parameter: nochange | toupper | tolower | common db names)) case - switch(case + 1, nochange, toupper, tolower) ### if (is.null(bulk_add)) ### bulk_add - .Call(RODBC:::C_RODBCCanAdd, attr(stat, handle_ptr)) structure(stat, class = RODBC, case = case, id = id, believeNRows = believeNRows, bulk_add = bulk_add, colQuote = colQuote, tabQuote = tabQuote, encoding = DBMSencoding, rows_at_time = rows_at_time) } __ 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] a first opinion on rattle
Received Fri 09 Jan 2009 7:49pm +1100 from Gabriel Ibarra: [...] I have installed rattle and is pretty intuitive and friendly. However, I miss some features of the original packages which cannot be invoked from rattle. For example, 'randomForest' package is??used but random??forests cannot be used in one of ??-in my opinion- most powerful aspects like regression. The same about SVM??. I would also say include RVM which the scientific literature suggest is mugh powerful than SVM. Anyway,??I must congratulate you and your team for providing the scientific community with rattle. Thanks for the most kind feedback Gabriel. It is appreciated. I am always open to suggestions. I should turn the randomForest regression option on. Thanks for suggesting that. RVM from kernlab is in the todo list now. Do you have some simple examples using rvm for regression? I should have a look to get that into Rattle. Help (i.e., code or just code examples) is always welcome :-) Thanks, Graham __ 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] shake rattle() and roll
Received Fri 09 Jan 2009 5:08pm +1100 from Dr Eberhard W Lisse: Graham, [...] Three Requests for Features for rattle(): would it not be nice to also have direct PostgreSQL (RdbiPgSQL) and MySQL (RMySQL) support and support for a SELECT statement? In other words, sometimes the database tables directly don't give you what you need and so you have to develop a (convoluted) SQL statement first. Would be nice if one could ratlle(convolutedSQL) :-)-O Hi Eberhard, Thanks for the feedback. I'm not a big user of ODBC and so much of that discussion (particularly re Mac) is beyond me. I've added the suggestion of directly supporting RdbiPgSQL and RMySQL to the Rattle todo list. Examples (i.e., R code) of loading data using these would be very helpful. So, basically, what is the sequence of R commands to get the data from either sources into a data frame in R (yes I could look it up but it would help, time-wise, to be spoon fed). The idea of adding the ability to have a SQl SELECT statement written in the GUI to extract just the data we want is good, and is on my todo list. Regards, Graham __ 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-pkgs] pmml 1.2.0 (predictive modelling markup language)
Version 1.2.0 of pmml has been released and is available from CRAN. The pmml package (http://rattle.togaware.com/pmml.html) is part of the Rattle data mining suite http://rattle.togaware.com. It generates representations of analytic models built in R using the open standard predictive modelling markup language (http://www.dmg.org/). PMML represents analytic models in an application independent way so that models developed in R can be loaded into other applications. Supported models include: arules, hclust, kmeans, ksvm, lm (and glm), nnet, rpart, and rsf. PMML consumers include Information Builders' C code generator (for deployment of R generated models within Web Focus on all of their supported platforms), Zementis' scoring engine or cloud computing application (ADAPA), Teradata's Warehouse Miner and IBM's DB2. A key update is the incorporation of a framework for dealing with data transformations. This is ongoing work. The pmml package is regularly updated and intermediate releases (between the more stable CRAN releases) are always available using: install.packages(pmml, repos=http://rattle.togaware.com;) Comments, bugs, suggestions are always welcome. Regards, Graham ___ R-packages mailing list r-packa...@r-project.org https://stat.ethz.ch/mailman/listinfo/r-packages __ 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-pkgs] Rattle 2.4.0 (Data Mining GUI using R)
Version 2.4.0 of Rattle has been released to CRAN. The rattle package (http://rattle.togaware.com) is a multi platform (GNU/Linux, Mac/OSX, MS/Windows) GTK based GUI for data mining (for exploring data and building descriptive and predictive models). It has undergone a lot of development over the past year. A companion book introducing data mining using Rattle is under development, with a draft available for review at: http://datamining.togaware.com/survivor/index.html. Rattle is regularly updated and intermediate releases (between the more stable CRAN releases) are always available using: install.packages(rattle, repos=http://rattle.togaware.com;) Rattle builds upon many R packages. Full installation instructions are provided at: http://datamining.togaware.com/survivor/Installing_Rattle.html Recent updates include: * Many bug fixes, GUI simplifications, improved colour usage (through the vcd package). * Streamlined handling of datasets (combined the Data and Select tabs), including support for CSV, ARFF, and ODBC data sources. * Addition of Test and Transforms tabs, with scripting support for transforms, both in building a model and in scoring. * Experimental support for automatic report generation using odfWeave (and hence, generation of OpenOffice documents). * Supported modelling includes: - Cluster (kmeans, hclust) - Association Rules (arules) - Linear Models (lm, glm) - Trees (rpart, party) - Neural Nets (nnet) - Support Vector Machines (ksvm) - Boosting (ada) - Random Forests (randomForest) * Supported data exploration tools include GGobi and PlayWith (latticist). * ROC curves, CostCurves, and many standard plots are supported. * Export to PMML is supported for many models, allowing R (and hence Rattle) models to be readily exported to other tools (see http://www.zementis.com) * The Business Intelligence vendor, Information Builders (http://ibi.com), will soon release RStat (sharing Rattle's open source code base) for data mining within WebFocus. Models built using RStat are deployable via PMML on all WebFocus hardware platforms (over 30 hardware and operating system platforms). A complete change log is available from http://code.google.com/p/rattle/source/list Comments, bugs, suggestions are always welcome. Regards, Graham ___ R-packages mailing list r-packa...@r-project.org https://stat.ethz.ch/mailman/listinfo/r-packages __ 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] Does file.info man page describe ctime corrrectly?
(R 2.8.0 on Debian GNU/Linux sid) ?file.info contains: mtime, ctime, atime: integer of class 'POSIXct': file modification, creation and last access times. This implies that ctime is file [...] creation [...] time Has R implemented ctime differently to Unix? I understand, on Linux at least, that ctime is the last change time (not the creation time). See man ls which says: -c [...] ctime (time of last modification of file status information) [...] However, testing seems to indicate file.info is returning the change time, not creation time (I'm not sure we can actually determine creation time of a file on Linux). Perhaps just a file.info documentation slip? Regards, Graham __ 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] Creating GUIs for R
Received Fri 10 Oct 2008 5:21am +1100 from Greg Snow: I am not involved in the RExcel project. I have just had some discussions with the people that are, so you should contact them for specific questions. I believe that this currently only works on windows, there was some mention of possibly expanding it to OpenOffice so that it would be cross platform, but I have no idea if that has made it past the good idea stage. I saw it demo'd on OpenOffice calc at UseR! 2008 as I recall --- it looked good. It was not yet released. For quick GUIs specialised for specific tasks, I use the Glade tool to graphically build the GUI - very quick and easy. Glade saves the GUI specs into XML. The RGtk2 package in R can read that XML file and display the GUI. The callbacks (i.e., the functionality called upon when you click a GUI button) are simply R functions. (Rattle is implemented this way.) Regards, Graham __ 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] playwith package does not work on Ubuntu?
Received Mon 30 Jun 2008 8:58pm +1000 from Karl Ove Hufthammer: Ingo Michaelis: I try to use the interactive plot package playwith on my machine (Ubuntu 8.04 64 bit / Gnome) using R version 2.7.1. My problem is that the widgets are not shown. I tried to plot the examples from the documentation, but it does not work anyway. FWIW, I have the same problem on a Fedora 8 system running KDE and the latest R Patched. Similarly Debian. The author is aware of the problem on GNU/Linux. Apparently works well on MS/Windows :-( The author is on holidays at the moment and has said he'll have a look at this once he returns. (Of course, source is available so anyone is welcome to have a go at fixing it.) It looks like a very nice package, once it works. Regards, Graham __ 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.