Hello, I stumbled upon nupic a couple of days ago, and started tinkering with it, with the aim of getting predictions on a function's output.
More specifically, I would like nupic to observe how an arbitrary function behaves, and then be able to know at time T what the value of the function is (or is expected to be), with or without observing it. Some extra context on what I want to do: scenario 1 -- to have a simple agent whose behavior is given by a function (like a simple state machine for example) which can be observed (and learned) by nupic, and then stop observations but still have "an idea" of what the agent is doing at any given time. scenario 2 -- at any time, I would like to be able to compute the prediction for the next N steps in the future, and only then (when I get to the future) compare my predictions (which I already have) to the real function output. >From what I read and tested, it looks like nupic could help me with what I need. However, I have tested the cpu example, but found it to predict only the current frame. I also searched in the mailing lists and other examples, but could not figure out the solution. How can I access future predictions ? Thank you ! Mike
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
