ECF is a C++ framework intended for application of any type of evolutionary computation - GA/GP, DE, ABC, Clonalg, PSO, GAn... - with an emphasis on genetic programming, GP. It offers simplicity for the end-user (parameterless usage: just provide the fitness function and go) and customization for experienced EC practicioners.
For assistance, check out project's home page for more information or use the forum or email project leader to get support.
Current features include:
- parameterless: genotype (individual structure) is the only mandatory parameter
- genetic algorithm genotypes (bitstring, binary encoded real values, floating point vectors, permutation vectors), genetic programming genotype (tree)
- individuals may contain any genotypes in any number (trees, bitstrings, permutation indices...)
- examples: function optimization, GP symbolic regression, GP artificial ant, TSP, onemax
- algorithms: steady state tournament, generational roulette-wheel, elimination, differential evolution (DE), particle swarm optimization (PSO), genetic annealing, random search
- parallel execution in many models (global paralel EA, distributed EA, hybrid parallel EA...) using MPI
- configurable environment: changing algorithm, genotypes and parameters without recompilation
- checkpointing and restoring mechanism, automated bacth runs
- some parameters: various termination criteria, population size and structure, crx and mutation operators selection and usage rate, migration