==== GAFT

A G\ enetic A\ lgorithm F\ ramework in py\ T\ hon

.. image:: https://travis-ci.org/PytLab/gaft.svg?branch=master :target: https://travis-ci.org/PytLab/gaft :alt: Build Status

.. image:: https://img.shields.io/codecov/c/github/PytLab/gaft/master.svg :target: https://codecov.io/gh/PytLab/gaft :alt: Codecov

.. image:: https://landscape.io/github/PytLab/gaft/master/landscape.svg?style=flat :target: https://landscape.io/github/PytLab/gaft/master :alt: Code Health

.. image:: https://img.shields.io/badge/python-3.5-green.svg :target: https://www.python.org/downloads/release/python-351/ :alt: platform

.. image:: https://img.shields.io/badge/pypi-v0.5.7-blue.svg :target: https://pypi.python.org/pypi/gaft/ :alt: versions

.. image:: https://readthedocs.org/projects/gaft/badge/?version=latest :target: https://gaft.readthedocs.io/en/latest/?badge=latest :alt: Documentation Status

Introduction

GAFT is a general Python Framework for genetic algorithm computation. It provides built-in genetic operators for target optimization and plugin interfaces for users to define your own genetic operators and on-the-fly analysis for algorithm testing.

GAFT is now accelerated using MPI parallelization interfaces. You can run it on your cluster in parallel with MPI environment.

Python Support

GAFT requires Python version 3.x (Python 2.x is not supported).

Installation

1. Via pip::

   pip install gaft

2. From source::

   python setup.py install

If you want GAFT to run in MPI env, please install mpi4py explicitly::

pip install mpi4py

See INSTALL.md <https://github.com/PytLab/gaft/blob/master/INSTALL.md>_ for more installation details.

Test

Run unit test::

python setup.py test

Quick start

1. Importing

.. code-block:: python

    from gaft import GAEngine
    from gaft.components import BinaryIndividual, Population
    from gaft.operators import RouletteWheelSelection, UniformCrossover, FlipBitMutation

    # Analysis plugin base class.
    from gaft.plugin_interfaces.analysis import OnTheFlyAnalysis

2. Define population

.. code-block:: python

indv_template = BinaryIndividual(ranges=[(0, 10)], eps=0.001)
population = Population(indv_template=indv_template, size=50)
population.init()  # Initialize population with individuals.

3. Create genetic operators

.. code-block:: python

    # Use built-in operators here.
    selection = RouletteWheelSelection()
    crossover = UniformCrossover(pc=0.8, pe=0.5)
    mutation = FlipBitMutation(pm=0.1)

4. Create genetic algorithm engine to run optimization

.. code-block:: python

engine = GAEngine(population=population, selection=selection,
                  crossover=crossover, mutation=mutation,
                  analysis=[FitnessStore])

5. Define and register fitness function

.. code-block:: python

    @engine.fitness_register
    def fitness(indv):
        x, = indv.solution
        return x + 10*sin(5*x) + 7*cos(4*x)

or if you want to minimize it, you can add a minimization decorator on it

.. code-block:: python

    @engine.fitness_register
    @engine.minimize
    def fitness(indv):
        x, = indv.solution
        return x + 10*sin(5*x) + 7*cos(4*x)

6. Define and register an on-the-fly analysis (optional)

.. code-block:: python

@engine.analysis_register
class ConsoleOutput(OnTheFlyAnalysis):
    master_only = True
    interval = 1
    def register_step(self, g, population, engine):
        best_indv = population.best_indv(engine.fitness)
        msg = 'Generation: {}, best fitness: {:.3f}'.format(g, engine.fmax)
        engine.logger.info(msg)

7. Run

.. code-block:: python

    if '__main__' == __name__:
        engine.run(ng=100)

8. Evolution curve

.. image:: https://github.com/PytLab/gaft/blob/master/examples/ex01/envolution_curve.png

9. Optimization animation

.. image:: https://github.com/PytLab/gaft/blob/master/examples/ex01/animation.gif

See `example 01 <https://github.com/PytLab/gaft/blob/master/examples/ex01/ex01.py>`_ for a one-dimension search for the global maximum of function `f(x) = x + 10sin(5x) + 7cos(4x)`

Global maximum search for binary function
-----------------------------------------

.. image:: https://github.com/PytLab/gaft/blob/master/examples/ex02/surface_animation.gif

See `example 02 <https://github.com/PytLab/gaft/blob/master/examples/ex02/ex02.py>`_ for a two-dimension search for the global maximum of function `f(x, y) = y*sin(2*pi*x) + x*cos(2*pi*y)`

Plugins
-------

You can define your own genetic operators for GAFT and run your algorithm test.

The plugin interfaces are defined in `/gaft/plugin_interfaces/ <https://github.com/PytLab/gaft/tree/master/gaft/plugin_interfaces>`_, you can extend the interface class and define your own analysis class or genetic operator class. The `built-in operators <https://github.com/PytLab/gaft/tree/master/gaft/operators>`_ and `built-in on-the-fly analysis <https://github.com/PytLab/gaft/tree/master/gaft/analysis>`_ can be treated as an official example for plugins development.

Blogs(Chinese Simplified)
-------------------------
- `GAFT-一个使用Python实现的遗传算法框架 <http://pytlab.github.io/2017/07/23/gaft-%E4%B8%80%E4%B8%AA%E5%9F%BA%E4%BA%8EPython%E7%9A%84%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95%E6%A1%86%E6%9E%B6/>`_

- `使用MPI并行化遗传算法框架GAFT <http://pytlab.github.io/2017/08/02/%E4%BD%BF%E7%94%A8MPI%E5%B9%B6%E8%A1%8C%E5%8C%96%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95/>`_

- `遗传算法中几种不同选择算子的比较 <http://pytlab.github.io/2017/09/19/%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95%E4%B8%AD%E5%87%A0%E7%A7%8D%E4%B8%8D%E5%90%8C%E9%80%89%E6%8B%A9%E7%AE%97%E5%AD%90%E7%9A%84%E6%AF%94%E8%BE%83/>`_

- `遗传算法中适值函数的标定与大变异算法 <http://pytlab.github.io/2017/09/23/%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95%E4%B8%AD%E9%80%82%E5%80%BC%E5%87%BD%E6%95%B0%E7%9A%84%E6%A0%87%E5%AE%9A%E4%B8%8E%E5%A4%A7%E5%8F%98%E5%BC%82%E7%AE%97%E6%B3%95/>`_

- `遗传算法框架GAFT优化小记 <http://pytlab.github.io/2017/10/08/%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95%E6%A1%86%E6%9E%B6GAFT%E4%BC%98%E5%8C%96%E5%B0%8F%E8%AE%B0/>`_

- `机器学习算法实践-Platt SMO和遗传算法优化SVM <http://pytlab.github.io/2017/10/15/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E7%AE%97%E6%B3%95%E5%AE%9E%E8%B7%B5-Platt-SMO%E5%92%8C%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95%E4%BC%98%E5%8C%96SVM/>`_

- `遗传算法框架GAFT已支持自定义个体编码方式 <http://pytlab.github.io/2018/03/07/%E9%81%97%E4%BC%A0%E7%AE%97%E6%B3%95%E6%A1%86%E6%9E%B6GAFT%E6%94%AF%E6%8C%81%E8%87%AA%E5%AE%9A%E4%B9%89%E4%B8%AA%E4%BD%93%E7%BC%96%E7%A0%81%E6%96%B9%E5%BC%8F/>`_

TODO
----
1. ✅ Parallelization 
2. ✅ Add more built-in genetic operators with different algorithms
3. 🏃 Add C++ backend(See `GASol <https://github.com/PytLab/GASol>`_)

Obtain a copy
-------------

The GAFT framework is distributed under the GPLv3 license and can be obtained from the GAFT git repository or PyPI 

- https://github.com/PytLab/gaft
- https://pypi.python.org/pypi/gaft/