David Ray
David Ray
@BoltzmannBrain [The C++ version](https://github.com/numenta/nupic.core/blob/master/src/nupic/algorithms/SpatialPooler.cpp#L1163) (which [calls to `addToWinners()` here](https://github.com/numenta/nupic.core/blob/master/src/nupic/algorithms/SpatialPooler.cpp#L1147)) seems to simply insert the winner into the vector in front of any winners with the same overlap score. So there's...
@subutai @rhyolight @BoltzmannBrain > The ordering amongst columns is forever fixed with a fixed random tie breaker. That's a really interesting point! Being "fixed" (i.e. deterministic) using an RNG with...
@subutai Wow! @vova616 has a very interesting idea! If I understand what he's saying correctly, the secondary sort (the tie-breaker) would become the distance between the presynaptic cell and the...
In a way. The problem only arises for "raw" SP use (i.e. use of the SP outside of the OPF or Network API), so in a way its exactly the...
> Wait, that's not what the issue description says. It says either the OPF or NAPI changes the `potentialRadius` within the SP. But whenever I initialize an SP object directly,...
@scottpurdy Also, what exactly happens with this parameter when `localInhibition=true` ? Does it get "tweaked" somewhere, or does the original value stand? What should this be set to, for localInhibition...
@rhyolight But I don't understand. Where would the user gain this knowledge of what to do? If there's no documentation for its affect on performance in different scenarios?
@rhyolight Ok, [I'm following your lead for how to treat this in HTM.Java](https://github.com/numenta/htm.java/issues/483) also... Also, it goes without saying that we are all on the edge of our seats, anxiously...
@rhyolight I think education about this is important. One look at this description for the `mapPotential()` method and you can see its obvious that this is not a "trivial" topic......