def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
# Create dict of arguments we will pass to the temporal memory class
args = copy.deepcopy(self.__dict__)
args["columnDimensions"] = (self.columnCount,)
# Create the TM instance
if self.monitor:
self._tm = createModel("extendedMixin", **args)
elif self.implementation == "cpp":
del args["columnCount"]
del args["formInternalConnections"]
del args["monitor"]
del args["implementation"]
del args["learningMode"]
del args["inferenceMode"]
del args["defaultOutputType"]
del args["_tm"]
self._tm = createModel("reversedExtendedCPP", **args)
else:
self._tm = createModel("extended", **args)
# numpy arrays we will use for some of the outputs
self.activeState = numpy.zeros(self._tm.numberOfCells())
self.previouslyPredictedCells = numpy.zeros(self._tm.numberOfCells())
def initialize(self):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
params = {
"columnDimensions": (self.columnCount,),
"basalInputDimensions": (self.basalInputWidth,),
"apicalInputDimensions": (self.apicalInputWidth,),
"cellsPerColumn": self.cellsPerColumn,
"activationThreshold": self.activationThreshold,
"initialPermanence": self.initialPermanence,
"connectedPermanence": self.connectedPermanence,
"minThreshold": self.minThreshold,
"maxNewSynapseCount": self.maxNewSynapseCount,
"permanenceIncrement": self.permanenceIncrement,
"permanenceDecrement": self.permanenceDecrement,
"predictedSegmentDecrement": self.predictedSegmentDecrement,
"formInternalBasalConnections": self.formInternalBasalConnections,
"learnOnOneCell": self.learnOnOneCell,
"maxSegmentsPerCell": self.maxSegmentsPerCell,
"maxSynapsesPerSegment": self.maxSynapsesPerSegment,
"seed": self.seed,
"checkInputs": self.checkInputs,
}
self._tm = createModel(self.implementation, **params)
def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
# Create dict of arguments we will pass to the temporal memory class
args = copy.deepcopy(self.__dict__)
args["columnDimensions"] = (self.columnCount, )
# Ensure we only pass in those args that are expected by this
# implementation. This is important for SWIG'ified classes, such as
# TemporalMemoryCPP, which don't take kwargs.
expectedArgs = getConstructorArguments(self.temporalImp)[0]
for arg in args.keys():
if not arg in expectedArgs:
args.pop(arg)
# Create the TM instance
self._tm = createModel(self.temporalImp, **args)
# numpy arrays we will use for some of the outputs
self.activeState = numpy.zeros(self._tm.numberOfCells())
self.previouslyPredictedCells = numpy.zeros(
self._tm.numberOfCells())
def initialize(self, dims, splitterMaps):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
params = {
"columnDimensions": (self.columnCount,),
"basalInputDimensions": (self.basalInputWidth,),
"apicalInputDimensions": (self.apicalInputWidth,),
"cellsPerColumn": self.cellsPerColumn,
"activationThreshold": self.activationThreshold,
"initialPermanence": self.initialPermanence,
"connectedPermanence": self.connectedPermanence,
"minThreshold": self.minThreshold,
"maxNewSynapseCount": self.maxNewSynapseCount,
"permanenceIncrement": self.permanenceIncrement,
"permanenceDecrement": self.permanenceDecrement,
"predictedSegmentDecrement": self.predictedSegmentDecrement,
"formInternalBasalConnections": self.formInternalBasalConnections,
"learnOnOneCell": self.learnOnOneCell,
"maxSegmentsPerCell": self.maxSegmentsPerCell,
"maxSynapsesPerSegment": self.maxSynapsesPerSegment,
"seed": self.seed,
"checkInputs": self.checkInputs,
}
self._tm = createModel(self.implementation, **params)
def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
# Create dict of arguments we will pass to the temporal memory class
args = copy.deepcopy(self.__dict__)
args["columnDimensions"] = (self.columnCount, )
self._tm = createModel(self.temporalImp, **args)
def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
# Create dict of arguments we will pass to the temporal memory class
args = copy.deepcopy(self.__dict__)
args["columnDimensions"] = (self.columnCount,)
self._tm = createModel(self.temporalImp, **args)
def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
# Create dict of arguments we will pass to the temporal memory class
args = copy.deepcopy(self.__dict__)
args["columnDimensions"] = (self.columnCount,)
# Create the TM instance
self._tm = createModel("extended", **args)
# numpy arrays we will use for some of the outputs
self.activeState = numpy.zeros(self._tm.numberOfCells())
self.previouslyPredictedCells = numpy.zeros(self._tm.numberOfCells())
def initialize(self):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
args = {
"columnDimensions": (self.columnCount, ),
"cellsPerColumn": self.cellsPerColumn,
"activationThreshold": self.activationThreshold,
"initialPermanence": self.initialPermanence,
"connectedPermanence": self.connectedPermanence,
"minThreshold": self.minThreshold,
"maxNewSynapseCount": self.maxNewSynapseCount,
"permanenceIncrement": self.permanenceIncrement,
"permanenceDecrement": self.permanenceDecrement,
"predictedSegmentDecrement": self.predictedSegmentDecrement,
"formInternalBasalConnections":
self.formInternalBasalConnections,
"learnOnOneCell": self.learnOnOneCell,
"maxSegmentsPerCell": self.maxSegmentsPerCell,
"maxSynapsesPerSegment": self.maxSynapsesPerSegment,
"seed": self.seed,
"checkInputs": self.checkInputs,
}
# Ensure we only pass in those args that are expected by this
# implementation. This is important for SWIG'ified classes, such as
# TemporalMemoryCPP, which don't take kwargs.
expectedArgs = getConstructorArguments(self.temporalImp)[0]
for arg in args.keys():
if not arg in expectedArgs:
args.pop(arg)
# Create the TM instance.
self._tm = createModel(self.implementation, **args)
# Carry some information to the next time step.
self.prevPredictiveCells = ()
self.prevActiveExternalCells = ()
self.prevActiveApicalCells = ()
def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
args = {
"columnDimensions": (self.columnCount,),
"cellsPerColumn": self.cellsPerColumn,
"activationThreshold": self.activationThreshold,
"initialPermanence": self.initialPermanence,
"connectedPermanence": self.connectedPermanence,
"minThreshold": self.minThreshold,
"maxNewSynapseCount": self.maxNewSynapseCount,
"permanenceIncrement": self.permanenceIncrement,
"permanenceDecrement": self.permanenceDecrement,
"predictedSegmentDecrement": self.predictedSegmentDecrement,
"formInternalBasalConnections": self.formInternalBasalConnections,
"learnOnOneCell": self.learnOnOneCell,
"maxSegmentsPerCell": self.maxSegmentsPerCell,
"maxSynapsesPerSegment": self.maxSynapsesPerSegment,
"seed": self.seed,
"checkInputs": self.checkInputs,
}
# Ensure we only pass in those args that are expected by this
# implementation. This is important for SWIG'ified classes, such as
# TemporalMemoryCPP, which don't take kwargs.
expectedArgs = getConstructorArguments(self.temporalImp)[0]
for arg in args.keys():
if not arg in expectedArgs:
args.pop(arg)
# Create the TM instance.
self._tm = createModel(self.implementation, **args)
# Carry some information to the next time step.
self.prevPredictiveCells = ()
self.prevActiveExternalCells = ()
self.prevActiveApicalCells = ()
def initialize(self, inputs, outputs):
"""
Initialize the self._tm if not already initialized. We need to figure out
the constructor parameters for each class, and send it to that constructor.
"""
if self._tm is None:
# Create dict of arguments we will pass to the temporal memory class
args = copy.deepcopy(self.__dict__)
args["columnDimensions"] = (self.columnCount,)
# Ensure we only pass in those args that are expected by this
# implementation. This is important for SWIG'ified classes, such as
# TemporalMemoryCPP, which don't take kwargs.
expectedArgs = getConstructorArguments(self.temporalImp)[0]
for arg in args.keys():
if not arg in expectedArgs:
args.pop(arg)
# Create the TM instance
self._tm = createModel(self.temporalImp, **args)
# numpy arrays we will use for some of the outputs
self.activeState = numpy.zeros(self._tm.numberOfCells())
self.previouslyPredictedCells = numpy.zeros(self._tm.numberOfCells())
def __init__(self,
inputWidth,
numActiveColumnsPerInhArea=40,
synPermProximalInc=0.1,
synPermProximalDec=0.001,
initialProximalPermanence=0.6,
columnDimensions=(2048,),
activationThreshold=13,
minThreshold=10,
initialPermanence=0.41,
connectedPermanence=0.50,
maxNewSynapseCount=20,
permanenceIncrement=0.10,
permanenceDecrement=0.10,
predictedSegmentDecrement=0.0,
maxSegmentsPerCell=255,
maxSynapsesPerSegment=255,
seed=42):
"""
This classes uses an ExtendedTemporalMemory internally to keep track of
distal segments. Please see ExtendedTemporalMemory for descriptions of
constructor parameters not defined below.
Parameters:
----------------------------
@param inputWidth (int)
The number of proximal inputs into this layer
@param numActiveColumnsPerInhArea (int)
Target number of active cells
@param synPermProximalInc (float)
Permanence increment for proximal synapses
@param synPermProximalDec (float)
Permanence decrement for proximal synapses
@param initialProximalPermanence (float)
Initial permanence value for proximal segments
"""
self.inputWidth = inputWidth
self.numActiveColumnsPerInhArea = numActiveColumnsPerInhArea
self.synPermProximalInc = synPermProximalInc
self.synPermProximalDec = synPermProximalDec
self.initialProximalPermanence = initialProximalPermanence
self.connectedPermanence = connectedPermanence
self.maxNewSynapseCount = maxNewSynapseCount
self.minThreshold = minThreshold
self.activeCells = set()
self._random = Random(seed)
# Create our own instance of extended temporal memory to handle distal
# segments.
self.tm = createModel(
modelName="extendedCPP",
columnDimensions=columnDimensions,
cellsPerColumn=1,
activationThreshold=activationThreshold,
initialPermanence=initialPermanence,
connectedPermanence=connectedPermanence,
minThreshold=minThreshold,
maxNewSynapseCount=maxNewSynapseCount,
permanenceIncrement=permanenceIncrement,
permanenceDecrement=permanenceDecrement,
predictedSegmentDecrement=predictedSegmentDecrement,
maxSegmentsPerCell=maxSegmentsPerCell,
maxSynapsesPerSegment=maxSynapsesPerSegment,
seed=seed,
learnOnOneCell=False,
)
# These sparse matrices will hold the synapses for each proximal segment.
#
# proximalPermanences - SparseMatrix with permanence values
# proximalConnections - SparseBinaryMatrix of connected synapses
self.proximalPermanences = SparseMatrix(self.numberOfColumns(),
inputWidth)
self.proximalConnections = SparseBinaryMatrix(inputWidth)
self.proximalConnections.resize(self.numberOfColumns(), inputWidth)
def __init__(self,
inputWidth,
lateralInputWidth,
numActiveColumnsPerInhArea=40,
synPermProximalInc=0.1,
synPermProximalDec=0.001,
initialProximalPermanence=0.6,
columnDimensions=(2048,),
minThresholdProximal=10,
activationThresholdDistal=13,
minThresholdDistal=10,
initialPermanence=0.41,
connectedPermanence=0.50,
maxNewProximalSynapseCount=20,
maxNewDistalSynapseCount=20,
permanenceIncrement=0.10,
permanenceDecrement=0.10,
predictedSegmentDecrement=0.0,
maxSegmentsPerCell=255,
maxSynapsesPerProximalSegment=255,
maxSynapsesPerDistalSegment=255,
seed=42):
"""
This classes uses an ExtendedTemporalMemory internally to keep track of
distal segments. Please see ExtendedTemporalMemory for descriptions of
constructor parameters not defined below.
Parameters:
----------------------------
@param inputWidth (int)
The number of proximal inputs into this layer
@param lateralInputWidth (int)
The number of lateral inputs into this layer
@param numActiveColumnsPerInhArea (int)
Target number of active cells
@param synPermProximalInc (float)
Permanence increment for proximal synapses
@param synPermProximalDec (float)
Permanence decrement for proximal synapses
@param initialProximalPermanence (float)
Initial permanence value for proximal segments
"""
self.inputWidth = inputWidth
self.lateralInputWidth = lateralInputWidth
self.numActiveColumnsPerInhArea = numActiveColumnsPerInhArea
self.synPermProximalInc = synPermProximalInc
self.synPermProximalDec = synPermProximalDec
self.initialProximalPermanence = initialProximalPermanence
self.connectedPermanence = connectedPermanence
self.maxNewProximalSynapseCount = maxNewProximalSynapseCount
self.maxNewDistalSynapseCount = maxNewDistalSynapseCount
self.minThresholdProximal = minThresholdProximal
self.minThresholdDistal = minThresholdDistal
self.maxSynapsesPerProximalSegment = maxSynapsesPerProximalSegment
self.activeCells = set()
self._random = Random(seed)
# Create our own instance of extended temporal memory to handle distal
# segments.
self.tm = createModel(
modelName="etm_cpp",
columnDimensions=columnDimensions,
basalInputDimensions=(lateralInputWidth,),
apicalInputDimensions=(),
cellsPerColumn=1,
activationThreshold=activationThresholdDistal,
initialPermanence=initialPermanence,
connectedPermanence=connectedPermanence,
minThreshold=minThresholdDistal,
maxNewSynapseCount=maxNewDistalSynapseCount,
permanenceIncrement=permanenceIncrement,
permanenceDecrement=permanenceDecrement,
predictedSegmentDecrement=predictedSegmentDecrement,
formInternalBasalConnections=False,
learnOnOneCell=False,
maxSegmentsPerCell=maxSegmentsPerCell,
maxSynapsesPerSegment=maxSynapsesPerDistalSegment,
seed=seed,
)
# These sparse matrices will hold the synapses for each proximal segment.
#
# proximalPermanences - SparseMatrix with permanence values
# proximalConnections - SparseBinaryMatrix of connected synapses
self.proximalPermanences = SparseMatrix(self.numberOfColumns(),
inputWidth)
self.proximalConnections = SparseBinaryMatrix(inputWidth)
self.proximalConnections.resize(self.numberOfColumns(), inputWidth)
