Monolithic .m files run demos on a couple of data sets.


Data sets
---------

binarydigits.txt  courtesy of Zoubin Ghahramani.

	bindigit2 visualises the data set, or look at bindigits.png
	NN runs nearest neighbours on this data set

binaryalphadigs.mat from:
	http://www.cs.toronto.edu/~roweis/data/binaryalphadigs.mat
	Courtesy of Sam Roweis.

	adapt bindigit2 or look at binaryalphadigs.jpg
	NN2 runs nearest neighbours on this data set


Fast methods for performing nearest neighbour searches on HUGE data sets exist.
See (eg)
	http://hunch.net/~jl/projects/cover_tree/cover_tree.html
	http://www.autonlab.org/autonweb/showProject/17/

