Hi Mark, > Is it possible to predict high and low within one nupic instance?
>From what I can tell it's currently impossible. And it's advised, like you mentioned, to run multiple instances, one for each desired field. I'll add to the questions: what is the limiting factor? OPF? or the classifier itself? (which is "artificial" part of the CLA, and could probably be reworked.) This should work if/once we have top-down "reconstruction" for that. Cheers, Mark -- Marek Otahal :o)
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
