Full sync of C++ temporal memory to Python temporal memory

src/nupic/support/group_by.py

@@ -0,0 +1,88 @@
+# ----------------------------------------------------------------------
+# Numenta Platform for Intelligent Computing (NuPIC)
+# Copyright (C) 2016, Numenta, Inc.  Unless you have an agreement
+# with Numenta, Inc., for a separate license for this software code, the
+# following terms and conditions apply:
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero Public License version 3 as
+# published by the Free Software Foundation.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+# See the GNU Affero Public License for more details.
+#
+# You should have received a copy of the GNU Affero Public License
+# along with this program.  If not, see http://www.gnu.org/licenses.
+#
+# http://numenta.org/licenses/
+# ----------------------------------------------------------------------
+
+from itertools import groupby
+
+
+def groupby2(*args):
+  """ An extension to groupby in itertools. Allows to walk across n sorted lists
+      with respect to their key functions and yields a tuple of n lists of the
+      members of the next *smallest* group.
+
+  @param args (list) a list of arguments alternating between sorted lists and
+                       their respective key functions. The lists should be
+                       sorted with respect to their key function.
+
+  @return (tuple) a n + 1 dimensional tuple, where the first element is the
+                    key of the group and the other n entries are lists of
+                    objects that are a member of the current group that is being
+                    iterated over in the nth list passed in. Note that this
+                    is a generator and a n+1 dimensional tuple is yielded for
+                    every group. If a list has no members in the current
+                    group, None is returned in place of a generator.
+
+  Notes: Read up on groupby here:
+         https://docs.python.org/dev/library/itertools.html#itertools.groupby
+
+"""
+  generatorList = [] # list of each list's (k, group) tuples
+
+  if len(args) % 2 == 1:
+    raise ValueError("Must have a key function for every list.")
+
+  # populate above lists
+  for i in xrange(0, len(args), 2):
+    listn = args[i]
+    fn = args[i + 1]
+    generatorList.append(groupby(listn, fn))
+
+  n = len(generatorList)
+
+  advanceList = [True] * n # start by advancing everyone.
+  nextList = [None] * n
+  # while all lists aren't exhausted walk through each group in order
+  while True:
+    for i in xrange(n):
+      if advanceList[i]:
+        try:
+          nextList[i] = generatorList[i].next()
+        except StopIteration:
+          nextList[i] = None
+
+    # no more values to process in any of the generators
+    if all(entry is None for entry in nextList):
+      break
+
+    # the minimum key value in the nextList
+    minKeyVal = min(nextVal[0] for nextVal in nextList
+                    if nextVal is not None)
+
+    # populate the tuple to return based on minKeyVal
+    retGroups = [minKeyVal]
+    for i in xrange(n):
+      if nextList[i] is not None and nextList[i][0] == minKeyVal:
+        retGroups.append(nextList[i][1])
+        advanceList[i] = True
+      else:
+        advanceList[i] = False
+        retGroups.append(None)
+
+    yield tuple(retGroups)

tests/integration/nupic/engine/network_creation_common.py

@@ -80,6 +80,7 @@
     "initialPerm": 0.21,
     "permanenceInc": 0.1,
     "permanenceDec": 0.1,
+    "predictedSegmentDecrement": .01,
     "globalDecay": 0.0,
     "maxAge": 0,
     "minThreshold": 9,

tests/integration/nupic/engine/temporal_memory_compatibility_test.py

@@ -31,16 +31,7 @@
 
 
 class TemporalMemoryCompatibilityTest(unittest.TestCase):
-  """Temporal Memory compatability tests between different implementations.
 
-  TODO: Test with predictedSegmentDecrement set to non-zero. See
-  https://github.com/numenta/nupic/issues/2999
-  """
-
-
-  @unittest.skip("There are slight differences between implementations that"
-                 "prevent this from passing still. See "
-                 "https://github.com/numenta/nupic/issues/2980")
   def testTMPyCpp(self):
     """
     Test compatibility between C++ and Python TM implementation.