similar to: SVM, feature selection and parameter tuning

Hi there I have an interesting problem: My csv file is of array dimensions [12,50], but it was saved the wrong way: there should be only 11 colums. What happens now if I read it into R is that the whole data set is shifted ( in the first row, the last column contains already the first value of the supposed second row and so on...) how can I tell R to switch after 11 read values to the next row,

Hello does anyone know how to save in a variable with a loop part. Im reading several csv files with read.table and would like to save with a loop: for (i in 0:9){ } should give read0, read1, read2, read3, read4, read5, read6, read7, read8, read9 st. like read(i)<-read.table("myfile.csv".....) I tried with paste("read", i, sep"") or to save as a vector with

Hello everyone, I have previously asked this question on the asterisk-video list, but I got directed here. I have a setup consisting of asterisknow beta4 (not sure if that is crucial) and a few clients all running X-Lite 3.0 (not eyebeam) on the local network. My computer has a USB-Camera installed, and now I would like to do some video calling with it, at least, so that the other user can

Hello all, I would like to perform SVM-RFE (Guyon et al. 2002) in R and have only found one implementation of this algorithm. The function belongs to the MCRestimate package but when I try to use it I encounter a problem - the function appears to be missing a required package or other function that I simply cannot find available anywhere. Here is my session info followed by a simple example

Hi folks, we have a problem with a win2003 DC and Samba. The authentification of users from the dc works fine, but when we added users from a forest trust in a active directory localgroup, samba don't find the users... I post this problem here: https://bugzilla.samba.org/show_bug.cgi?id=5245 Maybe you can help. Cheers Paul -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu

Dear all, Has someone implemented in R (or any other language) Knol DL, ten Berge JMF. Least-squares approximation of an improper correlation matrix by a proper one. Psychometrika, 1989, 54, 53-61. or any other similar algorithm? Best regards Jens Oehlschl?gel Background: I want to factanal() matrices of polychoric correlations which have negative eigenvalue. I coded Highham 2002

Hello to all, after noone bother my subject. I think I'm all alone making mistakes on this thing. But I should really know if it is possible to mange the administration of samba in an comfortable way. After all the years I worked with it I never did manage it to be in an sufficent manner. Is there a plan in the future to have a central administration within the samba package? --------

Hello, I'm currently using MCRestimate package and I have a question regarding the MCRestimate function. Here is my code: NestedCV.rf<-MCRestimate(eset, "Class", classificatin.fun="RF.wrap", variableSel.fun="varSel.highest.var", poss.parameters= list(var.numbers=c(100), mtry=c(10,50), cross.outer=10,cross.inner=10,cross.repeat=3) I'm pretty sure that I

Hello I am using support vector machine (from package kernlab) for a classification task (with RBF-Kernel). My data has dozens of variables and I need to identify which variables contribute most to the classification performance. What I did so far is comparing the classification performance (measured for example with the proportion of misclassified cases) of different sets of variables with

hi wine user group I try to run a model aircraft simulator (aerofly professional deluxe) with wine. the installation was a success. after that, I can start the simulator but it doesn't identify the usb game commander (at least a 4 channel usb device). but the kde desktop control center detect the right device IPACS Ikarus Gamecommander (/dev/input/js1). how I get support for this device

Hello list, I have trouble with a smb configuration that worked fine before I upgraded my samba to the latest stable release. I used samba 3.0.23c (Slackware11) and now I'm using samba 3.0.25b (Slackware12). The error/failure looks like this: I'm trying to connect to my HOME service and I fail. The log produces only this: [2007/08/13 10:29:33, 1]

hi, I would like to change the old windows 2000 domain structure into one samba 3 domain. the migration is to take place gradually for the departments. I had imagined to packing the samba pdc with another domain name into the same subnetz as windows pdc. Windows 2k DN = work.wurst.local Samba 3 DN = work.wurst.de IP Subnet = 192.168.2.0/24 (Samba and Windows) is it possible ? thank stefan --

Dear Group! I want to improve the firewall rules for SIP and I already compiled the linux kernel with additional SIP netfilter settings Now I found this on the internet: modprobe ip_conntrack_sip ip_nat_sip Set IPtables filter rules iptables -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT iptables -A INPUT -p udp --dport 5060 -j ACCEPT Set IPtables NAT rules iptables -A FORWARD -o

Hello, I did it from the same side on a suse 9.3. For me it worked with smpldap-useradd -m . But I had to recognize that some of the steps descriped there did not work on suse and I had to change different things in my ldap.conf and with pam and nss. Did yoe set smbpasswd -w yorldappassword and smbpasswd -a root perhaps you must have a usermap in smbusers root=administrator or just a net rpc

Hi! I've searched the web, but didn't find anything related to my strange problem: when I try to rename a file or a directory on a Samba share from my Windows-Client and I only change the case of a some characters it's ignored and the old filename is displayed again. Example: A directory "New Folder" should be renamed to "new Folder". Thats just ignored :-(. If I

Dear R Plots without par arguments do start not at zero (means, the box around is somwhere outside the specified plotrange). How to start really from zero, pe. basline y=0? every standart par works like this: x<-seq(1,10,0.1) y<-sin(x) plot(sin(x)) Thanks for help Marc -- Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten Browser-Versionen downloaden:

There are several people out there having problems with SNAT and Xen. I''m having a problem with DNAT and Xen and I think the reason for the problem could be exactly the same. I have actually tracked the problem down to http://lxr.mstier.de/linux/source/net/ipv4/route.c?v=2.6.16.28-xen302#1922 -- xenbr0 is the input device and it has no IP which is why everything coming in from that

Hi all - (Re my previous question to the list, I managed to generate an ROC plot for my SVM by ranking the data using the Decision.Values property. Thanks for your responses) I have now started tuning the SVM to get optimal parameters for the RBF kernel and I ran into a problem. Whatever parameter ranges I specify, I always get the same error values for all combinations of parameters it

Hello, if I try to tune my svm with the code: Tune <- tune.svm(Data.Train, Class.Train, type="C-classification", kernel="radial", gamma = 2^(-1:1), cost = 2^(2:4)) i get a windows Messagebox with a error in the application "Rgui.exe" and the message: "Die Anweisung in 0x6c48174d verweist auf Speicher 0x00000000. Der Vorgang "read" konnte nicht auf