Hi all, I would like know how one or more regions feed a higher region on the current NuPIC code. I know that all inputs coming from lower regions and sensors converge to a splittermap and then this splittermap is passed to SP. But my doubt is: how this splitter map is formed? (A) Is it formed by the concatenation of input maps of same size which each input map represents a region or sensor? (B) Or is it an array where each position is a alternation between bits and columns of each region or sensor? Or is it something else?
Bellow I give a rough draft of the I mean: Splitter map according to A: S1 | S1 | S1 | S1 | R1 | R1 | R1 | R1 | R2 | R2 | R2 | R2 S1 | S1 | S1 | S1 | R1 | R1 | R1 | R1 | R2 | R2 | R2 | R2 S1 | S1 | S1 | S1 | R1 | R1 | R1 | R1 | R2 | R2 | R2 | R2 S1 | S1 | S1 | S1 | R1 | R1 | R1 | R1 | R2 | R2 | R2 | R2 Splitter map according to B: S1 | R1 | R2 | S1 | R1 | R2 | S1 | R1 R2 | S1 | R1 | R2 | S1 | R1 | R2 | S1 R1 | R2 | S1 | R1 | R2 | S1 | R1 | R2 S1 | R1 | R2 | S1 | R1 | R2 | S1 | R1 R2 | S1 | R1 | R2 | S1 | R1 | R2 | S1 R1 | R2 | S1 | R1 | R2 | S1 | R1 | R2 Where S1 means "Sensor #1", R1 means "Region #1", R2 means "Region #2". Thanks in advance, PS: I'm sending this to general list because my acess to nupic-hackers is with problems. -- David Ragazzi MSc in Sofware Engineer (University of Liverpool) OS Community Commiter at Numenta.org -- "I think James Connolly, the Irish revolutionary, is right when he says that the only prophets are those who make their future. So we're not anticipating , we're working for it."
