similar to: gbm package: relationship between interaction.depth and number of features?

 Hello   I have questions about gbm package.  It seems we have to devide data to two part (training set and test set) for first.   1- trainig set for running of gbm function 2- test set for gbm.perf      is it rigth? I have 123 sample that I devided 100 for trainig and 23 for test.   So, parameter of cv.folds in gbm function is for what?   Thanks alot Azam       [[alternative HTML

Dear list, I would like to export a matrix to a TXT-File by using write.csv (not necessarily). Is there a way to add a header (with additional informations concerning the project) spanning multiple lines to this file before the actual data are listed up? Should look like this: date: filename: number of permutations: ------------ data (as a matrix) Any suggestions? Thnx in advance.

Suppose I have a small dataframe > aa Target Eaten ID 50 TPP 0 1 51 TPP 1 2 52 TPP 3 3 53 TPP 1 4 54 TPP 2 5 50.1 GPA 9 1 51.1 GPA 11 2 52.1 GPA 8 3 53.1 GPA 8 4 54.1 GPA 10 5 And I want to reshape it into ID TPP GPA 1 1 0 9 2 2 1 11 3 3 3 8 4 4 1 8 5 5 2 10 I realise that

I got Rmpi to compile with little difficulty, but had a tricky time setting the LD_LIBRARY_PATH to use the OpenMPI libs. I now get a different error when I try to load Rmpi > require(Rmpi) Loading required package: Rmpi librdmacm: couldn't read ABI version. librdmacm: assuming: 4 libibverbs: Fatal: couldn't read uverbs ABI version. CMA: unable to open

When I try to install Rmpi, it has a dependency rsprng which, as its description says: Provides interface to SPRNG 2.0 APIs I installed it on Mepis by installing the appropriate debs with minimal drama. In the rpm world of Fedora, no such luxury. Installing rsprng fails because: checking sprng.h usability... no checking sprng.h presence... no checking for sprng.h... no Cannot find sprng 2.0

Here is my code: mygbm<-gbm.fit(y=mytraindata[, 1], x=mytraindata[, -1], interaction.depth=4, shrinkage=0.001, n.trees=20000, bag.fraction=1, distribution="bernoulli") Here is the error: Error in gbm.fit(y = mytraindata[, 1], x = mytraindata[, -1], interaction.depth = 4, : The dataset size is too small or subsampling rate is too large: cRows*train.fraction*bag.fraction <=

Or, to avoid accusing you of lying. what you think is "vanilla" probably isn't. What exactly did you do? On Unix-likes, I would do something like this echo 'options(repos=list(CRAN="cran.r-project.org"));install.packages("Rcpp")' | R --vanilla (or maybe https://cloud.r-project.org is better...) -pd > On 2 Feb 2018, at 08:15 , Jeff Newmiller

I've noticed with certain datasets that when I try to do an anova and test for main effects and interaction for two explanatory variables, sometimes the main effect results are given but not the interaction results. For example, ex1 = aov(Score ~ var1*var2, data=myData) summary(ex1) gives me only the main effects for var1 and var2, but not the interaction. I also tried ex1 = aov(Score ~

Dear list members, i am currently want to install Rpy2 in a linux box which has R 2.4.0 installed RPy requries R 2.7.0 or above but i have no root previlleges so my question is how to install R 2.7.0 on my own directory? and replace the system installed R 2.4.0 when i input R command from the bash Shell? thanks for any reply. Best Regards! -- Daofeng Li,PhD Candidate China Agricultural

When i tried to install the hunspell package, I got this error message: Error: package ?Rcpp? 0.12.3 was found, but >= 0.12.12 is required by ?hunspell? So I set about installing a new version of Rcpp but I get this message: Error in unloadNamespace(pkg_name) : namespace ?Rcpp? is imported by ?dplyr? so cannot be unloaded How does one get around that? I tried installing Rcpp in a

On Wed, 17-May-2017 at 01:21PM +0200, Peter Dalgaard wrote: |> |> Anyways, you might want to |> |> a) move the discussion to R-devel |> b) include your platform (hardware, OS) and time zone info System: Host: MTA-V1-427894 Kernel: 3.19.0-32-generic x86_64 (64 bit gcc: 4.8.2) Desktop: KDE Plasma 4.14.2 (Qt 4.8.6) Distro: Linux Mint 17.3 Rosa Machine: System:

On Thu, 18-May-2017 at 05:46PM +0200, Martin Maechler wrote: |> ..... |> |> Being pretty "stretched" time wise currently, I'm happy for |> timezone-portable propositions to change the test. Meantime, anyone who lives where DST happpens in December who wants to get through the remaining tests can avoid this one by changing the line > stopifnot(length(fd) == 10,

The 3.2.4 release had two annoyances which we would rather not have in an "ultra-stable" release, designed to hang around for the duration of the 3.3 series. One was a relatively minor Makefile issue affecting system using R's bundled lzma library. The other, rather more serious, affected printing and formatting of POSIXlt objects, which would unpredictably get the Daylight Savings

I can use the gutenberg_download() function in the gutenbergr package on a computer that doeson't use a firewall, but on an almost identical installation that is behind a firewall, nothing happens, not even a time-out. Has anyone succeeded in using gutenberg_download() successfully with a firewall? I tried raising an issue at https://github.com/ropenscilabs/gutenbergr/issues/17 with no

I don't wish to install R by rpm. I need to know what Fedora rpms I need to install to give me the capability to install R using the tar.gz source file as I've done for years. On previous occasions when I've installed Fedora, I've used the DVD which has thousands of RPMs. Lately I've installed Fedora 15 from the Live CD which has a lot fewer and so a lot of necessary stuff

I don't wish to install R by rpm. I need to know what Fedora rpms I need to install to give me the capability to install R using the tar.gz source file as I've done for years. On previous occasions when I've installed Fedora, I've used the DVD which has thousands of RPMs. Lately I've installed Fedora 15 from the Live CD which has a lot fewer and so a lot of necessary stuff

On Fri, 02-Feb-2018 at 10:25AM +0100, peter dalgaard wrote: |> Or, to avoid accusing you of lying. what you think is "vanilla" |> probably isn't. What exactly did you do? On Unix-likes, I would do |> something like this |> echo 'options(repos=list(CRAN="cran.r-project.org"));install.packages("Rcpp")' | R --vanilla |> |> (or maybe

Your last statement is extremely unlikely to be true. The dplyr package should not be present in a vanilla environment, so there should be no such conflict. -- Sent from my phone. Please excuse my brevity. On February 1, 2018 11:00:01 PM PST, Patrick Connolly <p_connolly at slingshot.co.nz> wrote: >When i tried to install the hunspell package, I got this error >message: >

My CRAN mirror still says this: The latest release (Thursday 2016-03-10, Very Secure Dishes) R-3.2.4.tar.gz, read what's new in the latest version. Should that not be updated? Anyone who has not seen that post won't know to look further. On Wed, 16-Mar-2016 at 08:39PM +0000, Peter Dalgaard wrote: |> The 3.2.4 release had two annoyances which we would rather not have |> in

This has to do with your own timezone. If I run that code on my computer, both formats are correct. If I do this after Sys.setenv(TZ = "UTC") Then: > cbind(format(dlt), format(dct)) [,1] [,2] [1,] "2016-12-06 21:45:41" "2016-12-06 20:45:41" [2,] "2016-12-06 21:45:42" "2016-12-06 20:45:42" The reason for that, is that