Optimization framework

Under development!

Description:

Optimization framework implemented in Python.

UML diagram:

Usage:

Default config file is config.json.

./executor.py # loads configuration from default config file
./executor.py -c path/to/config.json # loads custom configuration

Algorithms:

• Genetic Algorithm - ga
• Pyramid Genetic Algorithm - pga
• Linkage Tree Genetic Algorithm - ltga
• Parameterless Pyramid Population Algorithm - p3
• Compact Genetic Algorithm - cga
• Pyramid Compact Genetic Algorithm - pcga
• Bayesan Optimization Algorithm - boa
• Pyramid Bayesan Optimization Algorithm - pboa

Problems:

• Boolean
• Capacited Vehicle Routing Problem
• Continuous optimization: Rastring, Schwefel, Griewank, Whitley, Rosenbrock
• Deceptive Trap
• Deceptive Step Trap
• Leading Ones
• Max Sat
• One Max
• Pipeline
• TSP:
• TSPLib: burma14, bays29, oliver30, brazil58, bier127, pr264, att532

Dependencies:

• numpy
• scipy

TODO:

• change places of config files
• rename Config to Context

Master thesis work

Marko Budiselić, Faculty of electrical engineering and computing, University of Zagreb, Croatia 2015