HHdEA - Hyper-heuristics for distributed Evolutionary Algorithms
To create jar files:
cd HHdEA
make It will generate HHdEA-1.0-SNAPSHOT-jar-with-dependencies.jar inside target folder.
Usage:
usage: java -cp <jar> br.ufpr.inf.cbio.hhdea.runner.Main [-a <algorithm>]
[-h] [-id <id>] [-M <methodology>] [-m <objectives>] [-o
<options...>] [-P <path>] [-p <problem>] [-s <seed>]
Execute a single independent run of the <algorithm> on a given <problem>.
-a,--algorithm <algorithm> set the algorithm to be executed:
NSGAII, MOEAD, MOEADD, ThetaDEA,
NSGAIII, SPEA2, SPEA2SDE, HypE, MOMBI2,
Traditional, HHdEA (default), <other>.If
<other> name is given, HHdEA will be
executed and the algorithm output name
will be <other>.
-h,--help print this message and exits.
-id <id> set the independent run id, default 0.
-M,--methodology <methodology> set the methodology to be used: NSGAIII
(default), MaF, Arion.
-m,--objectives <objectives> set the number of objectives to
<objectives> (default value is 3).
<problem> and <methodology> must be set
acordingly.
-o,--output <options...> enable some info output for
hyper-heuristics, space separated,
options include: fir firbin count.
-P,--output-path <path> directory path for output (if no path is
given experiment/ will be used.)
-p,--problem <problem> set the problem instance: DTLZ[1-7],
WFG[1-9], MinusDTLZ[1-7], MinusWFG[1-9],
MaF[1-15]; default is WFG1.<methodology>
must be set accordingly.
-s,--seed <seed> set the seed for JMetalRandom, default
System.currentTimeMillis()
Please report issues at https://github.com/fritsche/hhdea/issues