General features
- GP / GA / Differential Evolution support.
- Native support for concurrency.
- Essentially parameterless (very very few mandatory parameters).
- User programmable instruction set / termination criteria.
- Loading and saving of populations and genetic programs (checkpoint and restoring mechanism).
- Constraint handling.
- CSV / Weka XRFF dataset file format compatibility.
Algorithms
- Advanced linear genetic programming representation [1][2].
- ALPS (Age-Layered Population Structure) [9].
- Brood recombination [6][7]
- Cache for improving performance.
- Demetic grouping - Trivial Geography in Genetic Programming [3].
- Differential evolution [4][5][8].
- Gaussian Distribution for multi-class classification [13].
- Self-adaptive crossover operator [14].
- Steady state kernel.
- STGP (Strongly Typed Genetic Programming) [10].
- Team support [16].
Data
- Holdout validation.
- Stratified sampling.
Source
- Completely ISO C++20 compliant.
- Structured object oriented architecture.
- Strict const-correctness.
- Limited use of 3rd party libraries.
- Comprehensive, searchable API documentation generated from the source (Doxygen).