This document shows how to use
JCMsuite together with the high-level scripting language Matlab. This allows to setup complicated problems in a scripting-like manner and presents a simple and flexible way to define parameter dependent problems and to run parameter scans, optimizations or sensitivity analyses.
One major concept to combine
JCMsuite with Matlab is called Embedded Scripting. In a nutshell, the structure of the solver input is still based on
.jcm files. As an extension, these input files are allowed to contain placeholders (keys) for keyword substitution. Additionally, it is allowed to embed Matlab code snippets into the input files for looping over
.jcm blocks, or to dynamically update the key-value specification. A step-by-step tutorial is given here.
JCMsuite contains several numerical tools to perform efficient parameter optimizations, sensitivity analyses, and more. The domumentation of the Analysis and Optimization Toolkit explains how to use these tools within Matlab.
|||Matlab is a registered trademark of The MathWorks, Inc.|
|||Octave is mostly compatible to Matlab and is freely available from http://www.gnu.org/software/octave/ under the terms of the GNU General Public License (GPL) as published by the Free Software Foundation.|
Table Of Contents