OptimT v1.3.0

Updates:

• New module: PSO
• Add a single-objective standard PSO implementation, supporting both complie-time and run-time dimensional count, record and don't record fitness. Supports boosting using Eigen.
• Several chaotic devices ready for chaotic GA and chaotic PSO.
• Add functions to convert enumerations to strings.

Changes:

• GABase with recording is now a derived class from GABase without record.
• NSGA2 now uses hash checksum to examine whether the Pareto front has changed.

OptimTemplates v1.2.0

Updates:

• New algorithm implementation: NSGA2
• Several static assertions
• Chaos devices (prepared for CGA(chaotic genetic algorithm)
• Replace bool parameters in template with several enumeration types to make code more readable
• Initialize std::srand automatically with a time-based hash value.

Changes:

• Replace OPTIMT_MAKE_GLOBAL with OptimT_MAKE_GLOBAL
• Make randD global functions instead a static member of class OtGlobal. (old implementation reserved but marked deprecated)
• Replace qmake with cmake
• GABase are now more abstract

Bugs fixed:

• Fixed the bug that random numbers aren't random. (Be cautious to use std::random_device on Windows with MingW)

Since qmake has been replaced, there's no need to put an extra zipped file(with qmake project files removed) together with source code. Just download the soure code.

If any problems, don't hesitate to draw me an issue.

OptimTemplates v1.1.2-beta version

A beta version means that documents hasn't been updated.

OptimTemplates v1.1.0

New module:

1. Global
2. SimpleMatrix

Changes:

1. GABase are more abstract now to fit varities of genetic algorithms.

OptimTemplates v1.0.0

The first version of Ot!