A reusable implementation of the Simulated Annealing metaheuristic for optimization problems

Installation:

1. Run build.bat (Windows) or build.sh (unix) (creates dist/ folder if absent)
2. Change to dist/ and unpack compressed distribution
3. Change to unpacked distribution folder
4. run: sudo python setup.py install

This will install the simaneal package to Libs/site-packages

Files:

MANIFEST.in		-	Defines what will be in the final installer tarball
README			-	This file
runner			-	Python script to import this application's modules and launch
setup.py		-	distutil setup file
simanneal/		-	Simulated Annealing Package
	__init__.py	-	Used by distutils to find the package root, should be empty
	Annealer.py	-	Core implementation
	samples/	-	Sample package
		__init__.py		-	Used by distutils to find the package root, should be empty
		samplebase.py	-	Base class for samples, implements some simple defaults, children should only override what changes
		trivial.py		-	Sample of using Annealer to optimize a small one-dimensional array
		twodimensional.py	-	Sample of using Annealer to optimize a small two-dimensional array
		threedimensional.py	-	Guess what this does.
		visualizer.py		-	tkinter-based visualizer with threading for Annealer results (reusable for other data)
	tests/		-	Unit tests
		Annealer.py			- Unit tests for the core Annealer module
		twodimensional.py	- Unit tests for Sample, as implemented in twodimesional (yeah, unusual)

Interesting samples:
    - simanneal/samples/twodimensional.py - Command line run of simulated annealing in two dimensions using random starting data and interesting default values.
    - simanneal/samples/sample_visualizer.py - tkinter-based UI. Use 'random' to generate a random starting state or import some JSON data to start. the 'improve' button runs until the total energy of the system improves once. The 'anneal' button runs until a maximum energy threshold is reached. Requires tk to be installed on the host system.