Nope its the bug that Scott wrote the special Random method for. It was the bug re: the initPermanences() method and "connectedPct". https://github.com/numenta/nupic/issues/1321
Somehow I think this got forgotten about. But in reference to parameter potency - this is a vital one because it determines the distribution of initial connected synapses. "connectedPct" is the parameter in question though, not potentialPct like I thought. On Mon, Oct 27, 2014 at 5:55 PM, Matthew Taylor <[email protected]> wrote: > > On Wed, Oct 22, 2014 at 10:03 AM, cogmission1 . < > [email protected]> wrote: > >> Not to interrupt, but remember we determined that that setting >> (potentialPct) won't work until the fix is merged? >> > > David, are you talking about https://github.com/numenta/nupic/issues/1386? > I wasn't aware of any PR that addresses this issue yet. > > --------- > Matt Taylor > OS Community Flag-Bearer > Numenta > -- *We find it hard to hear what another is saying because of how loudly "who one is", speaks...*
