Home

Note: This is the wiki of my old trac/svn repo, and as yet only slightly updated. Coming soon...

Welcome to SGLib

SGLib is a toolbox for Matlab helping in the development of stochastic Galerkin and (to a lesser part as yet) stochastic collocation methods. More or less extensive documentation is available in toolbox itself. Here are provided only the:

Installation instructions

To have a smooth workflow with SGLib it is recommended to follow the steps outlined below:

• Go to your matlab development directory (i.e. where you put all your .m file for your stochastics project) and type in

  git clone git@github.com:ezander/sglib.git

This should retrieve all of the SGLib stuff and put it into directory sglib. (Here should come a description how to setup sglib as a subtree or submodule, what ever is preferred).

• Copy the file startup.m.template into your current directory as startup.m, if it's not already there.

  `cp sglib/startup.m.template startup.m`
  

If you already have a startup file you have to merge them (which shouldn't be too hard, since there is only one executable line in startup.m.template).

• Start matlab from the current directory (necessary for startup.m to be called automatically).

• You should be greeted by an SGLib welcome message (if not, there is either a bug in sglib or you may have a strange startup file or matlab shell script, then run sglib/sglib_startup manually)

• Run the testsuite to make sure everything is ok

  testsuite

• Start coding

Don't worry about overwriting anything in sglib. You can't. If you've messed something up there it's just local to you. You can just remove the complete sglib path and make 'git clone ...' to get a clean version.