Hi Nicholas, Improving this level of documentation is definitely a high priority for us this year. A couple of places you could look:
1) There's a simple example of using the temporal pooler directly here: examples/tp/hello_tp.py 2) Kevin is also working on a simple example of using the spatial pooler. The issue with the conversation is here: https://github.com/numenta/nupic/issues/654 and his current working code is here: https://github.com/lonesword/nupichelloworld/blob/master/helloworld.py 3) The hotgym client is probably the easiest one to use if you want an end to end system. 4) The python implementations of the algorithms are here if you want to directly modify them: py/nupic/research/spatial_pooler.py py/nupic/research/TP.py Hope this helps. --Subutai On Sat, Mar 1, 2014 at 7:09 AM, Nicholas Mitri <[email protected]> wrote: > Hello, > > I was wondering what the best way to approach learning the details of the > implementation was. > After getting the build ready and reading through some of the example > clients, what should a newcomer explore to get a better grasp of how the > implementation and the white paper differ? and possibly add their own > modifications at the algorithmic level? > > I'm interested in testing NUPIC from a machine learning perspective and > using it as part of a number of comparative studies with clustering and > learning algorithms. I haven't been able to get a firm handle on how > everything is implemented though. > > I'd appreciate any guidance in that respect :). > > Best, > Nicholas > _______________________________________________ > nupic mailing list > [email protected] > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
