[R] screeplot

Hi! I have a question concerning the screeplot function in R. What is the meaning of the colours in the plot. I applied princomp object to the screeplot funtion and it plotted the components against its variance, that's ok. But I have no idea about the different colouring of the boxes. I attach

[R] svm question

Hi! I have a questionaccording to classifing new examples from an already trained svm. I have tarined a svm (e1071 package) with a training set of 1526 examples. Now I have another data set with 2163 examples and I want to use the already trained svm for prediction: pred<-predict(a.svm,newdata

Re: [R] SVM question

Hi! I have another SVM question. I run and train a C-Classification using the SVM, works fine. Now I want to receive the output (the classified examples): pred <- fitted(model) where pred is then a list of the classification result of each input elememt i supposed. But if I compare the number o

[R] Problems with postscript output

Hi all! I have a little problem with saving plots to file. I use the command postscript() followed by the plotting command and a dev.off(). When I then look at the resulting image saved to disk, some of the axis labels are missing (see attached image). Is there a way to fix this. Yours, -- Fr

Re: [R] SVM question

On Mon, Mar 22, 2004 at 01:14:06PM +0100, David Meyer wrote: > >I have a question concerning the svm in the e1071 package. > >I trained the svm by a set of samples, doing a 10 cross validation. > >The summary function then prints out the total accuracy and single > >accuracies, >works fine. > > >M

[R] SVM question

Hi! I have a question concerning the svm in the e1071 package. I trained the svm by a set of samples, doing a 10 cross validation. The summary function then prints out the total accuracy and single accuracies, works fine. My question is then: Is it possible to get classification results per cros

[R] norm of complex number

Hi! I want o calc the norm of a complex number ( |c| where c is complex). Does the function abs() this ? Thanks, -- Frank G. Zoellner AG Angewandte Informatik Technische Fakult"at Universit"at Bielefeld phone: +49(0)521-106-2951 fax: +49(0)521-106-2992 email: [EMAIL PROTECTED] __

Re: [R] Receiver Operator Characteristic curve

On Thu, Mar 11, 2004 at 01:16:15PM -0500, XIAO LIU wrote: > Dear R-helpers: > > I want to calculate area under a Receiver Operator Characteristic curve. Where can > I find related functions? There is an ROC package within the Bioconductor Project that works with R. I use it, too. > > Thank yo

Re: [R] fft question

On Thu, Mar 11, 2004 at 06:14:02PM +0100, Martin Maechler wrote: > > Yes, manually, like > > fx <- fft(rep(x, 4)) > I think rep works on a vector but in my case x is a dataframe/matrix with the signal along the rows. Does rep work on dataframes ? Thanks, -- Frank G. Zoellner AG Angewand

[R] fft question

Hi! I am using the fft() function the base package to transform some 1d signal. If I use this standar fucntion I get a very huge first fourier coeficient. I think this dues to the handling of the borders of the signal. Usually in fft especially in image processing the signal is simulated to be co

[R] Changingvalues in data frame

Hi! I have a question concerning data frames and changing particular values in it. I have set up a data frame containing n rows of observations od dimension m, so in each row there is a vector of size m. Now I want to introduce a cut off to tha data. Therfore I caluclated the mean and variance

[R] SVM unbalanced classes

Hi! I am using R 1.8.1 and the svm of the e1071 package for classification. The problem is that I have unbalanced classes e.g. the first one is much bigger than the second one and therfore the svm is biased to the first class. If I manually adjust the class size the bias disappears. The question