Re: [R] tune an support vector machine

2013-12-07 Thread Uwe Bohne
Thank you very much, your proposal is one practical way to check for significant features. I tried to check for all combination in a loop, but unfortunately there is a problem with NA values. Maybe anybody has an idea. This is my expansion of the former code:

Re: [R] tune an support vector machine

2013-12-07 Thread Uwe Bohne
UPDATE This line of code will produce what i desired, i will check if the tuning of the svm works as i planed and post the solution asap l - apply(head(namen, -1), 1, function(x) reformulate(paste(na.omit(x), collapse = +), response = type)) l[[1]]

[R] wmf screen resolution problem!!!

2013-12-07 Thread Venkat Karthik
Dear colleagues, For the past two weeks we have been struggling to create a proper image with stable pixels, height width from R for various screen resolutions. We are trying to generate a wmf image with fixed pixels, fixed height fixed width. But the problem we are facing is that when the

[R] How to perform clustering without removing rows where NA is present in R

2013-12-07 Thread Gundala Viswanath
I have a data which contain some NA value in their elements. What I want to do is to **perform clustering without removing rows** where the NA is present. I understand that `gower` distance measure in `daisy` allow such situation. But why my code below doesn't work? __BEGIN__ # plot heat map

[R] ANOVA in R

2013-12-07 Thread Robin Mjelle
ID a_t1a_t2b_t1b_t2 CACCCGTAGAACCGACCTTGCG_mmu-miR-99b-5p15781941234810941 CACCCGTAGAACCGACCTTGC_mmu-miR-99b-5p4424265643839 CACCCGTAGAACCGACCTTG_mmu-miR-99b-5p544366253 CCGTAGAACCGACCTTGCG_mmu-miR-99b-5p263333157

Re: [R] ANOVA in R

2013-12-07 Thread Bert Gunter
I think you start by doing your homework: 1. Read An Inroduction to R (ships with R) or other R online tutorial. There are many good ones. 2. Use R's Help system: ?aov ?lm ?anova there will be relevant links in these docs that you should follow, especially to the use of formulas for model

[R] computational cost of the survpack or survival support vector machine package

2013-12-07 Thread Jieyue Li
Dear All, I'm trying using the survpack to apply support vector machines on survival analysis, but I find that it's very slow. I use a data frame with 1082 rows and 8 columns but it seems to run forever... Does anyone have any insight or suggestion to this or have better knowledge of another R

[R] combine glmnet and coxph (and survfit) with strata()

2013-12-07 Thread Jieyue Li
Dear All, I want to generate survival curve with cox model but I want to estimate the coefficients using glmnet. However, I also want to include a strata() term in the model. Could anyone please tell me how to have this strata() effect in the model in glmnet? I tried converting a formula with

Re: [R] wmf screen resolution problem!!!

2013-12-07 Thread Prof Brian Ripley
On 07/12/2013 12:46, Venkat Karthik wrote: Dear colleagues, For the past two weeks we have been struggling to create a proper image with stable pixels, height width from R for various screen resolutions. You cannot do that: it is a function of the format and how Microsoft's GDI works. We

[R] Why daisy() in cluster library failed to exclude NA when computing dissimilarity

2013-12-07 Thread Gundala Viswanath
Hi, According to daisy function from cluster documentation, it can compute dissimilarity when NA (missing) value(s) is present. http://stat.ethz.ch/R-manual/R-devel/library/cluster/html/daisy.html But why when I tried this code library(cluster) x - c(1.115,NA,NA,0.971,NA) y -