similar to: performance suggestion: sparse files

Hi, I have sparse (tridiagonal) matrices, and I use the Matrix package for handling them. For certain computations, I need to either set the first row to zero, or double the last row. I find that neither operation preserves sparsity, eg > m <- Diagonal(5) > m 5 x 5 diagonal matrix of class "ddiMatrix" [,1] [,2] [,3] [,4] [,5] [1,] 1 . . . . [2,] . 1

We have put a new package sparsenet on CRAN. Sparsenet fits regularization paths for sparse model selection via coordinate descent, using a penalized least-squares framework and a non-convex penalty. The package is based on our JASA paper Rahul Mazumder, Jerome Friedman and Trevor Hastie: SparseNet : Coordinate Descent with Non-Convex Penalties. (JASA 2011)

We have put a new package sparsenet on CRAN. Sparsenet fits regularization paths for sparse model selection via coordinate descent, using a penalized least-squares framework and a non-convex penalty. The package is based on our JASA paper Rahul Mazumder, Jerome Friedman and Trevor Hastie: SparseNet : Coordinate Descent with Non-Convex Penalties. (JASA 2011)

Hello people, I am in need for testing Marginal Homogeneity for sparse (more then 2X2) matrixes. After searching, what I found by now is that for more then 2 by 2 matrixes, one turns to "stuart maxwell tests" that are available in two packages: irr - see: stuart.maxwell.mh coin - see: mh_test But I couldn't find in the documentation how valid the results are for sparse matrixes,

Samba 3 svn repository have samba3 source code ,but I can't find samba3 document. Under /docs directory is empty. Where is samba3 document? -- --- Oota Toshiya --- oota at mail.linux.bs1.fc.nec.co.jp NEC Computers Software Operations Unit Shiba,Minato,Tokyo Linux Promotion Center Japan,Earth,Solar system (samba-jp Staff/postmaster,mutt-j/faqomatic admin,analog-jp

I have 2 samba server. both server is multi homed server. mostly same smb.conf. First server, run on FreeBSD 5-Stable,and this machine register two NetBIOS name to IP adress mapping to wins server. And this machine is m-node. Another server ,run on Linux,and this machine register only one NetBIOS name to IP adderss mapping to wins server. And this machine is h-node. I have a question. 1) Why

Dear all, I would like to compute power of a square non symmetric matrix. This is a part of a simulation study. Matrices are quite large (e.g., 900 by 900), and contains many 0 (more than 99 %). I have try the function mtx.exp of the Biodem package: library(Biodem) m <- matrix(0, 900, 900) i <- sample(1:900, 3000, replace = T) j <- sample(1:900, 3000, replace = T) for(x in 1:3000)

Oi, Bad timing of course, but the following is NOT an april's fools joke: I am working on a GTK+ interface to tinc, which should become an admin tool for the VPN that your tinc is a part of. Its name is pokey, from "pointy & clicky" -- and it is, of course, a reference to Pokey the Penguin. Right now it doesn't do much beside gathering and showing information to you. I

Hey gang, I was successful in recompiling my 2.4.20 kernel to support HDLC. I was successful in hooking up our T1 line into the zap card. I was successful in being able to ping equipment on the other end of the T1. I was unsuccessful in pinging the outside world from the other end of the T1. I've attached a cheezy image of the network. Here is the routing table: [root@asterisk root]# route

Update of /home/CVS/tinc/src/pokey In directory humbolt:/tmp/cvs-serv4765 Modified Files: pokey.glade Log Message: Commit diff test ***** Not enough context to create diff for file: s Tinc: Discussion list about the tinc VPN daemon Archive: http://mail.nl.linux.org/lists/ Tinc site: http://tinc.nl.linux.org/

Hello, I have next code: ;begin ; ModuleID = 'sample.lz' @.str1 = internal constant [20 x i8] c"\22hello, cruel world\22" ; <[20 x i8]*> [#uses=1] @.str4 = internal constant [9 x i8] c"\22hello, \22" ; <[9 x i8]*> [#uses=1] @.str7 = internal constant [7 x i8] c"\22heya!\22" ; <[7 x i8]*> [#uses=1]

Well, I have my 7940 up and running at home, and everything's working pretty much perfectly. So, it's time to find new things to break :-). The problem de jour is long-distance calling from work: I can use my cell phone, but I'm in an inside office on the 42nd floor and it's really hard to get good reception. So, I'm thinking about running a softphone on my Mac laptop

I am a beginner with ruby and was working through some basic examples from Onlamp where I create a basic controller and model, use the scaffold command, and edit my database.yml file to point it to the correct database. The problem I am having is that no matter what I do, I continue to get a mysql db connection error when I web to application. It seems that my connector is trying to connect to

Hi R-Users and Developers, Several months ago I made a request on Sourceforge to add the R/S - programming language to the _Trove_ categorization. ("The Trove is a means to convey basic metainformation about your project.") Today I got the following response of one of the sourceforge admins. <SNIP> SourceForge.net will consider the inclusion of a programming language within

Hi R-Users and Developers, Several months ago I made a request on Sourceforge to add the R/S - programming language to the _Trove_ categorization. ("The Trove is a means to convey basic metainformation about your project.") Today I got the following response of one of the sourceforge admins. <SNIP> SourceForge.net will consider the inclusion of a programming language within

Hi, I've been developing an application using CELT for some time and, for simplicity, have restricted frame sizes to powers of 2. During final testing I tested with a sound interface (the Phonic Firefly 302) whose ASIO drivers are very restictive in what they accept, and have about 6 allowable frame sizes < 512, a lot of which are odd numbers. However some are even numbered, so

> > so if it would work, replace shortname with short and short1? With all of this hokey-pokey surrounding licensing and mac addresses, I wonder if this outfit is actually still in compliance with the terms of their license for this software, whatever it may be? If the software licensed to run only on Machine X and Machine X has now been junked and replace by Machine Y, then isn't the

Buenas noches grupo. Saludos cordiales. He seguido en la búsqueda de una forma que me permita realizar la comparación de dos documentos pertenecientes a los años 2005 y 2013, y que pueda representar finalmente con wordcloud y con una table en la que las columnas sean los años de cada informe "2005" y "2013", y las filas sean las palabras con la frecuencia de cada una de ellas

Hello, I am trying to figure out how to make a bode plot (a.k.a. bode diagram) in ggplot2. An example of such a diagram can be found here: http://meweb.ecn.purdue.edu/~me475/ctm/freq/bode2.GIF The example above shows the gain and phase characteristic of a linear system. In my case, I would like to compare visually several systems on the same diagram. The data in the data frame is arranged

I've got more changes on my to-do list, but I want to ensure I'm making my changes against the right branch. * I'm working with trunk now. I don't think its TCP tunneling is as secure as the UDP tunneling. It looks like its IVs and HMACs are added and verified in send_udppacket and receive_udppacket. The TCP connection encrypts but doesn't have these anti-modification