Here is a list of all files with brief descriptions:

bpnet.hpp	This implements a plain backprop network
data.hpp	Contains formats for example data
genBoolMap.cpp	Generate a 2D grid (well, a table from which such a grid can be generated) of how many trials of UESMANN on every combination of binary boolean functions succeed
hinet.hpp	H-as-input network - only
mnist.hpp	Code for converting MNIST data into example sets
net.hpp	This is the abstract basic network class - the training methods are in each subclass
netFactory.hpp	I'm not a fan of factories, but here's one - this makes a network of the appropriate type which conforms to an example set, and is a namespace
netType.hpp	Contains integer enum for network types
obnet.hpp	Output blending network - only works with 2 h-levels, 0 and 1, and only with SGD
test.hpp	Useful stuff for testing
testBasic.cpp	Basic tests of underlying functionality, or things which only take a short time to run!
testSaveLoad.cpp	Tests of loading and saving. These work by generating random networks, running them, and load/save cycling them to see if the params are the same
testTrainBasic.cpp	Tests of basic training
testTrainBooleans.cpp	Simple boolean tests for modulation
uesnet.hpp	This file contains the implementation of the UESMANN network itself - at least, those parts which are different from a standard Rumelhart/Hinton/Williams MLP