Home
| Browse documentation online |
|
Quick download: 0.5.3 release 0.5.3 examples 0.5.3 PDF/TR |
|
Don't forget to join and contribute your questions on the MAILING LIST |
NEWS: METSlib 0.5.3 (2011-05-05):
|
METSlib is a metaheuristic modeling framework and optimization toolkit in modern C++ released as !Free/Libre/Open Source Software.
Model and algorithms are modular: any search algorithm can be applied to the same model. On the other hand no assumption is made on the model, you can work on any problem type: timetabling, assignment problems, vehicle routing, bin-packing and so on.
Once you have implemented your model in the problem framework, the library makes easy testing different Tabu Search strategies or even different algorithms (Simulated Annealing or other local search based algorithms) with a few lines of code.
METSlib hides the nuts and bolts of the algorithms from your code, you are only supposed to code the problem logic, not the search algorithm, this results in clearer code, which is also easier to read: your search strategy can be better understood and modified with less effort.
The toolkit implements the basics of some metaheuristic algorithms:
- Simple local search
- Simulated Annealing (with linear, exponential and custom cooling schedule)
- Last but not least, Tabu Search.
It's really easy to code other metaheuristics based on the previous ones:
- Random Restart Local Search
- Variable Neighborhood Search
- Iterated Local Search
METSlib was tested with GCC 4.3.x and 4.4.x, but written with portability in mind, so, even if no build file for specific IDEs are provided, it should be simple to add one: the only requirement is a fairly recent C++ compiler with support for the TR1 extension. The correct way to build the library is to use the provided configure script (the script needs a Unix system or, on Windows, the CygWin compatibility layer).
To use any algorithm you must implement an objective function, a neighborhood exploration strategy (move manager) and the actual moves.
If you choose Tabu Search metaheuristic you are free to use some of the already implemented termination criteria, tabu lists, and aspiration criteria, or you can implement you own specialized versions.
If you opt for Simulated Annealing you can use one of the provided cooling schedules or implement your own.
This OO library was inspired by the OTS library released by the Coin-OR project. It was started by me in 2006 during the period of my master thesis with invaluable support from people of the Global Optimization Laboratory, and refined in the following years.
METSlib is Open Source and is covered by the GNU GPL v3 license or, at you option, by the EPL-1.0. Other licenses can be considered by the author if needed. Feel free to join the mailing list, create issues, send hints, and contribution.
- Are you using METSlib in your project? I'll be pleased to know what you are using METSlib for and to have your feedback. Write me at <mirko (dot) maischberger (at) gmail (dot) com>.
- Have you spotted a typo or a poor english sentence? Please report it to me or correct it yourself in pure wiki spirit.
- Would you like to suggest a missing feature? Join the mailing list, discuss the feature and, possibly, provide code ;).
Version 0.5.3 is a public release of METSlib on COIN-OR.
-
New (alpha) document describing METSlib usage (please send me your comments on this document)
-
Please join the mailing list for support requests and bug reporting.
- Release repository of older, buggy and possibly useless versions of the library.
- Migration Guide from 0.4.x to 0.5.x
-- Mirko Maischberger