We are proud to announce the release of DEAP 0.9.0.
You can download a copy of this release at the following web page.
DEAP (Distributed Evolutionary Algorithms in Python) is a novel evolutionary
computation framework for rapid prototyping and testing of ideas. Its design
departs from most other existing frameworks in that it seeks to make algorithms
explicit and data structures transparent, as opposed to the more common black
box type of frameworks.
This release includes :
- major overhaul of the genetic programming with significant speed increase;
- new state of the art operators to control bloat in GP;
- ability to do GP with arbitrary Python objects, i.e. NumPy vectors and matrices;
- several new examples from diverse fields;
- organization of the examples by category;
- examples are now compatible with Python 2 and 3 out of the box;
- and several other changes.
A major change for DEAP is that from 0.9.0, the easy distribution module DTM
will be replaced by SCOOP (Scalable COncurrent Operations in Python), a
distributed task module allowing concurrent parallel programming on various
environments, from heterogeneous grids to supercomputers. This new project is
led by Yannick Hold-Geoffroy (@yannickhold) in close association with DEAP's
developers. You can download the last release at the following web page.
Your feedback and comments are welcome at http://goo.gl/2HiO1
deap-users at googlegroups dot com. You can also follow us on Twitter @deapdev,
and on our blog http://deapdev.wordpress.com/.
François-Michel De Rainville
Laboratoire de vision et systèmes numériques
Département de génie électrique et génie informatique
Quebec City (Quebec), Canada