def constructTM(self, columnCount, basalInputSize, apicalInputSize, cellsPerColumn, initialPermanence, connectedPermanence, minThreshold, sampleSize, permanenceIncrement, permanenceDecrement, predictedSegmentDecrement, activationThreshold, seed): params = { "columnDimensions": (columnCount,), "basalInputDimensions": (basalInputSize,), "apicalInputDimensions": (apicalInputSize,), "cellsPerColumn": cellsPerColumn, "initialPermanence": initialPermanence, "connectedPermanence": connectedPermanence, "minThreshold": minThreshold, "maxNewSynapseCount": sampleSize, "permanenceIncrement": permanenceIncrement, "permanenceDecrement": permanenceDecrement, "predictedSegmentDecrement": predictedSegmentDecrement, "activationThreshold": activationThreshold, "seed": seed, "learnOnOneCell": False, "formInternalBasalConnections": False, } self.tm = ExtendedTemporalMemory(**params)
def initialize(self): rangePadding = abs(self.inputMax - self.inputMin) * 0.2 minVal = self.inputMin - rangePadding maxVal = (self.inputMax + rangePadding if self.inputMin != self.inputMax else self.inputMin + 1) numBuckets = 130.0 resolution = max(0.001, (maxVal - minVal) / numBuckets) self.valueEncoder = RandomDistributedScalarEncoder(resolution, w=41, seed=42) self.encodedValue = np.zeros(self.valueEncoder.getWidth(), dtype=np.uint32) self.timestampEncoder = DateEncoder(timeOfDay=( 21, 9.49, )) self.encodedTimestamp = np.zeros(self.timestampEncoder.getWidth(), dtype=np.uint32) inputWidth = self.valueEncoder.getWidth() self.sp = SpatialPooler( **{ "globalInhibition": True, "columnDimensions": [2048], "inputDimensions": [inputWidth], "potentialRadius": inputWidth, "numActiveColumnsPerInhArea": 40, "seed": 1956, "potentialPct": 0.8, "boostStrength": 0.0, "synPermActiveInc": 0.003, "synPermConnected": 0.2, "synPermInactiveDec": 0.0005, }) self.spOutput = np.zeros(2048, dtype=np.float32) self.etm = ExtendedTemporalMemory( **{ "activationThreshold": 13, "cellsPerColumn": 1, "columnDimensions": (2048, ), "basalInputDimensions": (self.timestampEncoder.getWidth(), ), "initialPermanence": 0.21, "maxSegmentsPerCell": 128, "maxSynapsesPerSegment": 32, "minThreshold": 10, "maxNewSynapseCount": 20, "permanenceDecrement": 0.1, "permanenceIncrement": 0.1, "seed": 1960, "checkInputs": False, }) learningPeriod = math.floor(self.probationaryPeriod / 2.0) self.anomalyLikelihood = anomaly_likelihood.AnomalyLikelihood( claLearningPeriod=learningPeriod, estimationSamples=self.probationaryPeriod - learningPeriod, reestimationPeriod=100)
def initialize(self): rangePadding = abs(self.inputMax - self.inputMin) * 0.2 minVal = self.inputMin - rangePadding maxVal = (self.inputMax + rangePadding if self.inputMin != self.inputMax else self.inputMin + 1) numBuckets = 130.0 resolution = max(0.001, (maxVal - minVal) / numBuckets) self.valueEncoder = RandomDistributedScalarEncoder(resolution, w=41, seed=42) self.encodedValue = np.zeros(self.valueEncoder.getWidth(), dtype=np.uint32) self.timestampEncoder = DateEncoder(timeOfDay=(21,9.49,)) self.encodedTimestamp = np.zeros(self.timestampEncoder.getWidth(), dtype=np.uint32) inputWidth = self.valueEncoder.getWidth() self.sp = SpatialPooler(**{ "globalInhibition": True, "columnDimensions": [2048], "inputDimensions": [inputWidth], "potentialRadius": inputWidth, "numActiveColumnsPerInhArea": 40, "seed": 1956, "potentialPct": 0.8, "boostStrength": 0.0, "synPermActiveInc": 0.003, "synPermConnected": 0.2, "synPermInactiveDec": 0.0005, }) self.spOutput = np.zeros(2048, dtype=np.float32) self.etm = ExtendedTemporalMemory(**{ "activationThreshold": 13, "cellsPerColumn": 1, "columnDimensions": (2048,), "basalInputDimensions": (self.timestampEncoder.getWidth(),), "initialPermanence": 0.21, "maxSegmentsPerCell": 128, "maxSynapsesPerSegment": 32, "minThreshold": 10, "maxNewSynapseCount": 20, "permanenceDecrement": 0.1, "permanenceIncrement": 0.1, "seed": 1960, "checkInputs": False, }) learningPeriod = math.floor(self.probationaryPeriod / 2.0) self.anomalyLikelihood = anomaly_likelihood.AnomalyLikelihood( claLearningPeriod=learningPeriod, estimationSamples=self.probationaryPeriod - learningPeriod, reestimationPeriod=100 )
class ExtendedTMCPP_ApicalTiebreakSequencesTests( ApicalTiebreakSequencesTestBase, unittest.TestCase): """ Run the apical tiebreak sequence tests on the C++ ExtendedTemporalMemory. """ def constructTM(self, columnCount, apicalInputSize, cellsPerColumn, initialPermanence, connectedPermanence, minThreshold, sampleSize, permanenceIncrement, permanenceDecrement, predictedSegmentDecrement, activationThreshold, seed): params = { "apicalInputDimensions": (apicalInputSize, ), "columnDimensions": (columnCount, ), "cellsPerColumn": cellsPerColumn, "initialPermanence": initialPermanence, "connectedPermanence": connectedPermanence, "minThreshold": minThreshold, "maxNewSynapseCount": sampleSize, "permanenceIncrement": permanenceIncrement, "permanenceDecrement": permanenceDecrement, "predictedSegmentDecrement": predictedSegmentDecrement, "activationThreshold": activationThreshold, "seed": seed, "learnOnOneCell": False, "formInternalBasalConnections": True, } self.tm = ExtendedTemporalMemory(**params) def compute(self, activeColumns, apicalInput, learn): activeColumns = sorted(activeColumns) apicalInput = sorted(apicalInput) # Use depolarizeCells + activateCells rather than tm.compute so that # getPredictiveCells returns predictions for the current timestep. self.tm.depolarizeCells(activeCellsExternalApical=apicalInput, learn=learn) self.tm.activateCells(activeColumns, reinforceCandidatesExternalApical=apicalInput, growthCandidatesExternalApical=apicalInput, learn=learn) def reset(self): self.tm.reset() def getActiveCells(self): return self.tm.getActiveCells() def getPredictedCells(self): return self.tm.getPredictiveCells()
class ExtendedTMCPP_SequenceMemoryTests(SequenceMemoryTestBase, unittest.TestCase): """ Run the sequence memory tests on the C++ ExtendedTemporalMemory. """ def constructTM(self, columnCount, cellsPerColumn, initialPermanence, connectedPermanence, minThreshold, sampleSize, permanenceIncrement, permanenceDecrement, predictedSegmentDecrement, activationThreshold, seed): params = { "columnDimensions": (columnCount,), "cellsPerColumn": cellsPerColumn, "initialPermanence": initialPermanence, "connectedPermanence": connectedPermanence, "minThreshold": minThreshold, "maxNewSynapseCount": sampleSize, "permanenceIncrement": permanenceIncrement, "permanenceDecrement": permanenceDecrement, "predictedSegmentDecrement": predictedSegmentDecrement, "activationThreshold": activationThreshold, "seed": seed, "learnOnOneCell": False, } self.tm = ExtendedTemporalMemory(**params) def compute(self, activeColumns, learn): # Use depolarizeCells + activateCells rather than tm.compute so that # getPredictiveCells returns predictions for the current timestep. self.tm.depolarizeCells(learn=learn) self.tm.activateCells(sorted(activeColumns), learn=learn) def reset(self): self.tm.reset() def getActiveCells(self): return self.tm.getActiveCells() def getPredictedCells(self): return self.tm.getPredictiveCells()
class DistalTimestamps1CellPerColumnDetector(AnomalyDetector): """The 'numenta' detector, with the following changes: - Use pure Temporal Memory, not the classic TP that uses backtracking. - Don't spatial pool the timestamp. Pass it in as distal input. - 1 cell per column. - Use w=41 in the scalar encoding, rather than w=21, to make up for the lost timestamp input to the spatial pooler. """ def __init__(self, *args, **kwargs): super(DistalTimestamps1CellPerColumnDetector, self).__init__(*args, **kwargs) self.valueEncoder = None self.encodedValue = None self.timestampEncoder = None self.encodedTimestamp = None self.activeExternalCells = [] self.prevActiveExternalCells = [] self.sp = None self.spOutput = None self.etm = None self.anomalyLikelihood = None def getAdditionalHeaders(self): """Returns a list of strings.""" return ["raw_score"] def initialize(self): rangePadding = abs(self.inputMax - self.inputMin) * 0.2 minVal = self.inputMin - rangePadding maxVal = (self.inputMax + rangePadding if self.inputMin != self.inputMax else self.inputMin + 1) numBuckets = 130.0 resolution = max(0.001, (maxVal - minVal) / numBuckets) self.valueEncoder = RandomDistributedScalarEncoder(resolution, w=41, seed=42) self.encodedValue = np.zeros(self.valueEncoder.getWidth(), dtype=np.uint32) self.timestampEncoder = DateEncoder(timeOfDay=(21,9.49,)) self.encodedTimestamp = np.zeros(self.timestampEncoder.getWidth(), dtype=np.uint32) inputWidth = self.valueEncoder.getWidth() self.sp = SpatialPooler(**{ "globalInhibition": True, "columnDimensions": [2048], "inputDimensions": [inputWidth], "potentialRadius": inputWidth, "numActiveColumnsPerInhArea": 40, "seed": 1956, "potentialPct": 0.8, "boostStrength": 0.0, "synPermActiveInc": 0.003, "synPermConnected": 0.2, "synPermInactiveDec": 0.0005, }) self.spOutput = np.zeros(2048, dtype=np.float32) self.etm = ExtendedTemporalMemory(**{ "activationThreshold": 13, "cellsPerColumn": 1, "columnDimensions": (2048,), "basalInputDimensions": (self.timestampEncoder.getWidth(),), "initialPermanence": 0.21, "maxSegmentsPerCell": 128, "maxSynapsesPerSegment": 32, "minThreshold": 10, "maxNewSynapseCount": 20, "permanenceDecrement": 0.1, "permanenceIncrement": 0.1, "seed": 1960, "checkInputs": False, }) learningPeriod = math.floor(self.probationaryPeriod / 2.0) self.anomalyLikelihood = anomaly_likelihood.AnomalyLikelihood( claLearningPeriod=learningPeriod, estimationSamples=self.probationaryPeriod - learningPeriod, reestimationPeriod=100 ) def handleRecord(self, inputData): """Returns a tuple (anomalyScore, rawScore).""" self.valueEncoder.encodeIntoArray(inputData["value"], self.encodedValue) self.timestampEncoder.encodeIntoArray(inputData["timestamp"], self.encodedTimestamp) self.prevActiveExternalCells = self.activeExternalCells self.activeExternalCells = self.encodedTimestamp.nonzero()[0] self.sp.compute(self.encodedValue, True, self.spOutput) activeColumns = self.spOutput.nonzero()[0] activeColumnsSet = set(activeColumns.tolist()) prevPredictedColumns = set(self.etm.columnForCell(cell) for cell in self.etm.getPredictiveCells()) rawScore = (len(activeColumnsSet - prevPredictedColumns) / float(len(activeColumns))) anomalyScore = self.anomalyLikelihood.anomalyProbability( inputData["value"], rawScore, inputData["timestamp"]) logScore = self.anomalyLikelihood.computeLogLikelihood(anomalyScore) self.etm.compute(activeColumns, activeCellsExternalBasal=self.activeExternalCells, reinforceCandidatesExternalBasal=self.prevActiveExternalCells, growthCandidatesExternalBasal=self.prevActiveExternalCells) return (logScore, rawScore)
class DistalTimestamps1CellPerColumnDetector(AnomalyDetector): """The 'numenta' detector, with the following changes: - Use pure Temporal Memory, not the classic TP that uses backtracking. - Don't spatial pool the timestamp. Pass it in as distal input. - 1 cell per column. - Use w=41 in the scalar encoding, rather than w=21, to make up for the lost timestamp input to the spatial pooler. """ def __init__(self, *args, **kwargs): super(DistalTimestamps1CellPerColumnDetector, self).__init__(*args, **kwargs) self.valueEncoder = None self.encodedValue = None self.timestampEncoder = None self.encodedTimestamp = None self.activeExternalCells = [] self.prevActiveExternalCells = [] self.sp = None self.spOutput = None self.etm = None self.anomalyLikelihood = None def getAdditionalHeaders(self): """Returns a list of strings.""" return ["raw_score"] def initialize(self): rangePadding = abs(self.inputMax - self.inputMin) * 0.2 minVal = self.inputMin - rangePadding maxVal = (self.inputMax + rangePadding if self.inputMin != self.inputMax else self.inputMin + 1) numBuckets = 130.0 resolution = max(0.001, (maxVal - minVal) / numBuckets) self.valueEncoder = RandomDistributedScalarEncoder(resolution, w=41, seed=42) self.encodedValue = np.zeros(self.valueEncoder.getWidth(), dtype=np.uint32) self.timestampEncoder = DateEncoder(timeOfDay=( 21, 9.49, )) self.encodedTimestamp = np.zeros(self.timestampEncoder.getWidth(), dtype=np.uint32) inputWidth = self.valueEncoder.getWidth() self.sp = SpatialPooler( **{ "globalInhibition": True, "columnDimensions": [2048], "inputDimensions": [inputWidth], "potentialRadius": inputWidth, "numActiveColumnsPerInhArea": 40, "seed": 1956, "potentialPct": 0.8, "boostStrength": 0.0, "synPermActiveInc": 0.003, "synPermConnected": 0.2, "synPermInactiveDec": 0.0005, }) self.spOutput = np.zeros(2048, dtype=np.float32) self.etm = ExtendedTemporalMemory( **{ "activationThreshold": 13, "cellsPerColumn": 1, "columnDimensions": (2048, ), "basalInputDimensions": (self.timestampEncoder.getWidth(), ), "initialPermanence": 0.21, "maxSegmentsPerCell": 128, "maxSynapsesPerSegment": 32, "minThreshold": 10, "maxNewSynapseCount": 20, "permanenceDecrement": 0.1, "permanenceIncrement": 0.1, "seed": 1960, "checkInputs": False, }) learningPeriod = math.floor(self.probationaryPeriod / 2.0) self.anomalyLikelihood = anomaly_likelihood.AnomalyLikelihood( claLearningPeriod=learningPeriod, estimationSamples=self.probationaryPeriod - learningPeriod, reestimationPeriod=100) def handleRecord(self, inputData): """Returns a tuple (anomalyScore, rawScore).""" self.valueEncoder.encodeIntoArray(inputData["value"], self.encodedValue) self.timestampEncoder.encodeIntoArray(inputData["timestamp"], self.encodedTimestamp) self.prevActiveExternalCells = self.activeExternalCells self.activeExternalCells = self.encodedTimestamp.nonzero()[0] self.sp.compute(self.encodedValue, True, self.spOutput) activeColumns = self.spOutput.nonzero()[0] activeColumnsSet = set(activeColumns.tolist()) prevPredictedColumns = set( self.etm.columnForCell(cell) for cell in self.etm.getPredictiveCells()) rawScore = (len(activeColumnsSet - prevPredictedColumns) / float(len(activeColumns))) anomalyScore = self.anomalyLikelihood.anomalyProbability( inputData["value"], rawScore, inputData["timestamp"]) logScore = self.anomalyLikelihood.computeLogLikelihood(anomalyScore) self.etm.compute( activeColumns, activeCellsExternalBasal=self.activeExternalCells, reinforceCandidatesExternalBasal=self.prevActiveExternalCells, growthCandidatesExternalBasal=self.prevActiveExternalCells) return (logScore, rawScore)