matrix to vector by row of by col in R

m <- matrix(....) 

as.vector(m)   # by column 

as.vector(t(m)) # by row 

Grep Recursively Through Single File Extension

grep --include=*.py -lir "delete" .

from http://www.commandlinefu.com/commands/view/9046/grep-recursively-through-single-file-extension

grep tab

After some searching I found out the answer is to not use \t but to use Ctrl-v and then the <tab> key to put the tab character directly in the command. Odd. So it looks like:

 grep -R "   $" *

where the tab character is inserted as Ctrl-v and then Tab.

copied from http://mandrivausers.org/index.php?/topic/34048-grepping-for-tabs/

See Where a Package is Installed on Ubuntu

copied from

http://www.howtogeek.com/howto/ubuntu/see-where-a-package-is-installed-on-ubuntu/

 

See Where a Package is Installed on Ubuntu

Once you use the apt-get utility to install a package, sometimes it seems to disappear into nowhere. You know it's installed, you just have no idea where.

If you know the name of the executable, you can use the which command to find the location of the binary, but that doesn't give you information on where the supporting files might be located.

There's an easy way to see the locations of all the files installed as part of the package, using the dpkg utility.

dpkg -L <packagename>

use gs to reduce pdf size somehow

gs -sDEVICE=pdfwrite -dCompatibilityLevel=1.4 -dPDFSETTINGS=/default -dNOPAUSE -dQUIET -dBATCH -sOutputFile=output.pdf a.pdf

from http://www.ubuntugeek.com/ubuntu-tiphowto-reduce-adobe-acrobat-file-size-from-command-line.html

vim virtualedit

set virtualedit=onemore so that some mapping still works when the cursor is at the end of file. 

jags data format and openbugs data format

From JAGS manual:

There is no need to transpose matrices and arrays when transferring data between R and JAGS,

since JAGS stores the values of an array in "column major" order, like R and FORTRAN (i.e.

filling the left-hand index first).

If you have an S-style data file for WinBUGS and you wish to convert it for JAGS, then

use the command bugs2jags, which is supplied with the coda package.

From http://mathstat.helsinki.fi/openbugs/Manuals/ModelSpecification.html#RepeatedStructures

Data can be in an S-Plus format (see most of the examples ) or, for data in arrays, in rectangular format.

Missing values are represented as NA.

The whole array must be specified in the file - it is not possible just to specify selected components. Any parts of the array you do not want to specify must be filled with NAs.

All variables in a data file must be defined in a model, even if just left unattached to the rest of the model. 

S-Plus format: This allows scalars and arrays to be named and given values in a single structure headed by key-word list. There must be no space after list.

For example, in the Rats example, we need to specify a scalar xbar , dimensions N and T , a vector x and a two-dimensional array Y with 30 rows and 5 columns. This is achieved using the following format:

         list(
               xbar = 22, N = 30, T = 5,
               x = c(8.0, 15.0, 22.0, 29.0, 36.0),
               Y = structure(
                     .Data = c(
                           151, 199, 246, 283, 320,
                           145, 199, 249, 293, 354,
                           ...................
                           ...................
                           137, 180, 219, 258, 291,
                           153, 200, 244, 286, 324),
                     .Dim = c(30, 5)
                     )
         )

OpenBUGS reads data into an array by filling the right-most index first, whereas the S-Plus program fills the left-most index first. Hence OpenBUGS reads the string of numbers c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10) into a 2 * 5 dimensional matrix in the order   
   [i, j]th element of matrix      value
      [1, 1]      1
      [1, 2]      2
      [1, 3]      3
      .....      ..
      [1, 5]      5
      [2, 1]      6
      .....      ..
      [2, 5]      10

whereas S-Plus reads the same string of numbers in the order

   [i, j]th element of matrix      value
      [1, 1]      1
      [2, 1]      2
      [1, 2]      3
      .....      ..
      [1, 3]      5
      [2, 3]      6
      .....      ..
      [2, 5]      10

Hence the ordering of the array dimensions must be reversed before using the S-Plus dput function to create a data file for input into OpenBUGS .

For example, consider the 2 * 5 dimensional matrix

         1   2   3   4   5
         6   7   8   9   10

This must be stored in S-Plus as a 5 * 2 dimensional matrix:

> M
[,1] [,2]
[1,] 1 6
[2,] 2 7
[3,] 3 8
[4,] 4 9
[5,] 5 10

The S-Plus command

> dput(list(M=M), file="matrix.dat")

will then produce the following data file

list(M = structure(.Data = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10), .Dim=c(5,2))

Edit the .Dim statement in this file from .Dim=c(5,2) to .Dim=c(2,5). The file is now in the correct format to input the required 2 * 5 dimensional matrix into OpenBUGS .

Now consider a 3 * 2 * 4 dimensional array

         1   2   3   4
         5   6   7   8

         9   10   11   12
         13   14   15   16

         17   18   19   20
         21   22   23   24

This must be stored in S-Plus as the 4 * 2 * 3 dimensional array:

> A
, , 1
[,1] [,2]
[1,] 1 5
[2,] 2 6
[3,] 3 7
[4,] 4 8

, , 2
[,1] [,2]
[1,] 9 13
[2,] 10 14
[3,] 11 15
[4,] 12 16

, , 3
[,1] [,2]
[1,] 17 21
[2,] 18 22
[3,] 19 23
[4,] 20 24

The command

> dput(list(A=A), file="array.dat")

will then produce the following data file

list(A = structure(.Data = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 1 6, 17, 18, 19, 20, 21, 22, 23, 24), .Dim=c(4,2,3)) 

Edit the .Dim statement in this file from .Dim=c(4,2,3) to .Dim=c(3,2,4). The file is now in the correct format to input the required 3 * 2 * 4 dimensional array into OpenBUGS in the order