DEAP (Distributed Evolutionary Algorithms in Python) is a free, open-source and cross-platform evolutionary computation framework that allows you to quickly perform prototyping processes and testing of ideas.
DEAP aims to make algorithms easy to understand and data structures transparent, and implements intuitive parallelism, eliminating the unnecessary implementation details.
Detailed instructions on how to install and use the DEAP utility on your Mac are available HERE.
DEAP is cross-platform and it works on Mac OS X, Windows and Linux.
Here are some key features of "DEAP":
· Provides an environment to quickly implement algorithm
· Allow for straightforward parallelization
· Substantial set of real-world examples
Requirements:
· Python
What's New in This Release: [ read full changelog ]
Bug fixes:
· The class creator._numpy_array could not be pickled/unpicled properly, therefore when unpickling an object inheriting from numpy.ndarray and created with the creator, attributes were missing.
· algorithms.varAnd was not working properly with odd population size;
Examples:
· coev_symbreg now evaluate every individual at each generation;
· pso_speciation and pso_multiswarm are functional again with Python 2;
New features:
· Re-introduction of gp.mutShrink operator, now also working with STGP;
· Introduction of gp.graph, a function that return the essential to draw a PrimitveTree;
Documentation:
· New tutorial on genetic programming;
· More visible deprecation warning;
· More links to SCOOP.
Find us on Google+
