BacktrackingTM topDownCompute confusion

rhyolight · May 24, 2017, 10:50pm

I’m working on backtracking_tm.py API docs, and there’s this function:

numenta/nupic/blob/8b48371c5c5a08c7813c247233694ec7ed12bb96/src/nupic/algorithms/backtracking_tm.py#L2449-L2462


def topDownCompute(self, topDownIn=None):
  """
  Top-down compute - generate expected input given output of the TM


  @param topDownIn top down input from the level above us


  @returns best estimate of the TM input that would have generated bottomUpOut.
  """
  # For now, we will assume there is no one above us and that bottomUpOut is
  # simply the output that corresponds to our currently stored column
  # confidences.


  # Simply return the column confidences
  return self.columnConfidences()

Which ignores the topDownIn method parameter and simply returns self.columnConfidences() instead. Then you have that function:

github.com

numenta/nupic/blob/8b48371c5c5a08c7813c247233694ec7ed12bb96/src/nupic/algorithms/backtracking_tm.py#L2435-L2446


def columnConfidences(self, cellConfidences=None):
  """
  Compute the column confidences given the cell confidences. If
  None is passed in for cellConfidences, it uses the stored cell confidences
  from the last compute.


  @param cellConfidences Cell confidences to use, or None to use the
                         the current cell confidences.


  @returns Column confidence scores
  """
  return self.colConfidence['t']

Which ignores the cellConfidences method parameter as well.

What’s the story with this? Seems like it needs cleanup.

subutai · May 25, 2017, 8:25pm

Legacy code - thumbs up for removing it as long as it is not used anywhere critical (I doubt it.)

rhyolight · May 25, 2017, 8:40pm

Thanks Subutai, I’ll have a PR soon.

Topic		Replies	Views
Backtracking TM and Cell4.cpp don't work NuPIC	6	791	July 18, 2018
Same parameter get different result when using BacktrackingTMCPP NAB	4	588	September 16, 2019
Python 3 Migration: Backtracking tests NuPIC python	7	993	February 15, 2018
Python 3 Migration: TM's not matching NuPIC python	0	669	January 31, 2018
Biological Constraints NuPIC Community Fork	4	668	April 10, 2019

BacktrackingTM topDownCompute confusion

Related topics