Download EvA2

Current Release

The most recent release of EvA2 can be found here. The base package contains the core framework and GUI classes. The EvA2 Problems package adds additional optimization problems.

Name Links Content
EvA2 Base Package Binary [jar],  source [zip] The EvA2 base package with all optimization strategies, data types and basic operators used. This is all you need for most applications, including sophisticated own implementations. Use 'java -jar EvA2.jar' or 'java -cp EvA2.jar eva2.gui.Main' to start EvA2 from a console.
EvA2 Problems Package Binary [jar],   source [zip] Additional benchmark problems, such as a part of the CEC 2005 benchmark suite as well as additional multi-modal and multi-criterial benchmarks not contained in the base package. Use 'java -cp EvA2.jar:EvA2Problems.jar eva2.gui.Main' to start EvA2 using the problem package from a console. Note that the path separator character (':' in GNU/Linux) is platform-dependent.
EvA2 Matlab Interface m-Files [tgz] Archive containing the Matlab interface files only. They are contained in the resources folder of the base package as well.
SBTOOLBOX2 update to EvA2 v2.034 [tbz2] Decompress into SBTOOLBOX2/tools/optimization/auxiliary/EvA2 for usage. Thanks to Milan Mitrovic.

Older Versions

Older versions of EvA2 may be found here.

Former Release

Name Size Content
JavaEvA.jar 2.705 kb The JavaEvA Optimization toolbox as downloadable jar version. You can download this version to your PC and start JavaEvA by double clicking the JavaEvA.jar file or by using the 'java -jar JavaEvA.jar' or 'java -cp JavaEvA javaeva.client.EvAClient' command from a console. 5.209 kb This is an exemplary framework, which illustrates how to implement your on optimization problem and optimize it using the JavaEvA package. This example includes the JavaEvA.jar, examplary problem implementations and ant targets to start the JavaEvA package and to start the instances of the optimzation problems. 5.016 kb The source code of the JavaEvA Optimization toolbox as downloadable zip.


Title Type Size
JavaEvA: A Java based framework for Evolutionary Algorithms [BibTeX] Technical Report (PDF, 2005) 1,744 kB