def createNetwork(dataSource):
"""Create the Network instance.
The network has a sensor region reading data from `dataSource` and passing
the encoded representation to an Identity Region.
:param dataSource: a RecordStream instance to get data from
:returns: a Network instance ready to run
"""
network = Network()
# Our input is sensor data from the gym file. The RecordSensor region
# allows us to specify a file record stream as the input source via the
# dataSource attribute.
network.addRegion("sensor", "py.RecordSensor",
json.dumps({"verbosity": _VERBOSITY}))
sensor = network.regions["sensor"].getSelf()
# The RecordSensor needs to know how to encode the input values
sensor.encoder = createEncoder()
# Specify the dataSource as a file record stream instance
sensor.dataSource = dataSource
# CUSTOM REGION
# Add path to custom region to PYTHONPATH
# NOTE: Before using a custom region, please modify your PYTHONPATH
# export PYTHONPATH="<path to custom region module>:$PYTHONPATH"
# In this demo, we have modified it using sys.path.append since we need it to
# have an effect on this program.
sys.path.append(os.path.dirname(os.path.abspath(__file__)))
# Add custom region package to network
Network.registerRegionPackage("custom_region")
# Create a custom region
network.addRegion("identityRegion", "py.IdentityRegion", json.dumps(I_PARAMS))
# Link the Identity region to the sensor input
network.link("sensor", "identityRegion", "UniformLink", "",
srcOutput="sourceOut", destInput="in")
network.initialize()
return network
def createNetwork(dataSource):
"""Create the Network instance.
The network has a sensor region reading data from `dataSource` and passing
the encoded representation to an SPRegion. The SPRegion output is passed to
a TPRegion.
:param dataSource: a RecordStream instance to get data from
:returns: a Network instance ready to run
"""
network = Network()
# Our input is sensor data from the gym file. The RecordSensor region
# allows us to specify a file record stream as the input source via the
# dataSource attribute.
network.addRegion("sensor", "py.RecordSensor",
json.dumps({"verbosity": _VERBOSITY}))
sensor = network.regions["sensor"].getSelf()
# The RecordSensor needs to know how to encode the input values
sensor.encoder = createEncoder()
# Specify the dataSource as a file record stream instance
sensor.dataSource = dataSource
# Create the spatial pooler region
SP_PARAMS["inputWidth"] = sensor.encoder.getWidth()
network.addRegion("spatialPoolerRegion", "py.SPRegion", json.dumps(SP_PARAMS))
# Link the SP region to the sensor input
network.link("sensor", "spatialPoolerRegion", "UniformLink", "")
network.link("sensor", "spatialPoolerRegion", "UniformLink", "",
srcOutput="resetOut", destInput="resetIn")
network.link("spatialPoolerRegion", "sensor", "UniformLink", "",
srcOutput="spatialTopDownOut", destInput="spatialTopDownIn")
network.link("spatialPoolerRegion", "sensor", "UniformLink", "",
srcOutput="temporalTopDownOut", destInput="temporalTopDownIn")
# Add the TPRegion on top of the SPRegion
# TODO: Needs TMRegion
network.addRegion("temporalMemoryRegion", "py.TPRegion",
json.dumps(TP_PARAMS))
network.link("spatialPoolerRegion", "temporalMemoryRegion", "UniformLink", "")
network.link("temporalMemoryRegion", "spatialPoolerRegion", "UniformLink", "",
srcOutput="topDownOut", destInput="topDownIn")
# Register TPRegion since we aren't in nupic
curDirectory = os.path.dirname(os.path.abspath(__file__))
# directory containing the union pooler directory is 2 directories above this file
unionTemporalPoolerDirectory = os.path.split((os.path.split(curDirectory))[0])[0]
sys.path.append(unionTemporalPoolerDirectory)
Network.registerRegionPackage("union_temporal_pooling")
# Add the TPRegion on top of the TPRegion
temporal = network.regions["temporalMemoryRegion"].getSelf()
UP_PARAMS["inputWidth"] = temporal.getOutputElementCount("bottomUpOut")
network.addRegion("unionTemporalPoolerRegion", "py.TemporalPoolerRegion", json.dumps(UP_PARAMS))
network.link("temporalMemoryRegion", "unionTemporalPoolerRegion", "UniformLink", "",
srcOutput="activeCells", destInput="activeCells")
network.link("temporalMemoryRegion", "unionTemporalPoolerRegion", "UniformLink", "",
srcOutput="predictedActiveCells", destInput="predictedActiveCells")
network.initialize()
spatial = network.regions["spatialPoolerRegion"].getSelf()
# Make sure learning is enabled (this is the default)
spatial.setParameter("learningMode", 1, True)
# We want temporal anomalies so disable anomalyMode in the SP. This mode is
# used for computing anomalies in a non-temporal model.
spatial.setParameter("anomalyMode", 1, False)
# Enable topDownMode to get the predicted columns output
temporal.setParameter("topDownMode", 1, True)
# Make sure learning is enabled (this is the default)
temporal.setParameter("learningMode", 1, True)
# Enable inference mode so we get predictions
temporal.setParameter("inferenceMode", 1, True)
temporal.setParameter("computePredictedActiveCellIndices", 1, True)
union = network.regions["unionTemporalPoolerRegion"].getSelf()
# Make sure learning is enabled (this is the default)
union.setParameter("learningMode", 1, True)
return network
def createNetwork(dataSource):
"""Create the Network instance.
The network has a sensor region reading data from `dataSource` and passing
the encoded representation to an SPRegion. The SPRegion output is passed to
a TPRegion.
:param dataSource: a RecordStream instance to get data from
:returns: a Network instance ready to run
"""
network = Network()
# Our input is sensor data from the gym file. The RecordSensor region
# allows us to specify a file record stream as the input source via the
# dataSource attribute.
network.addRegion("sensor", "py.RecordSensor",
json.dumps({"verbosity": _VERBOSITY}))
sensor = network.regions["sensor"].getSelf()
# The RecordSensor needs to know how to encode the input values
sensor.encoder = createEncoder()
# Specify the dataSource as a file record stream instance
sensor.dataSource = dataSource
# Create the spatial pooler region
SP_PARAMS["inputWidth"] = sensor.encoder.getWidth()
network.addRegion("spatialPoolerRegion", "py.SPRegion", json.dumps(SP_PARAMS))
# Link the SP region to the sensor input
network.link("sensor", "spatialPoolerRegion", "UniformLink", "")
network.link("sensor", "spatialPoolerRegion", "UniformLink", "",
srcOutput="resetOut", destInput="resetIn")
network.link("spatialPoolerRegion", "sensor", "UniformLink", "",
srcOutput="spatialTopDownOut", destInput="spatialTopDownIn")
network.link("spatialPoolerRegion", "sensor", "UniformLink", "",
srcOutput="temporalTopDownOut", destInput="temporalTopDownIn")
# Add the TPRegion on top of the SPRegion
# TODO: Needs TMRegion
network.addRegion("temporalMemoryRegion", "py.TPRegion",
json.dumps(TP_PARAMS))
network.link("spatialPoolerRegion", "temporalMemoryRegion", "UniformLink", "")
network.link("temporalMemoryRegion", "spatialPoolerRegion", "UniformLink", "",
srcOutput="topDownOut", destInput="topDownIn")
# Register UPRegion since we aren't in nupic
curDirectory = os.path.dirname(os.path.abspath(__file__))
# directory containing the union pooler directory is 2 directories above this file
unionPoolerDirectory = os.path.split((os.path.split(curDirectory))[0])[0]
sys.path.append(unionPoolerDirectory)
Network.registerRegionPackage("union_pooling")
# Add the UPRegion on top of the TPRegion
temporal = network.regions["temporalMemoryRegion"].getSelf()
UP_PARAMS["inputWidth"] = temporal.getOutputElementCount("bottomUpOut")
network.addRegion("unionPoolerRegion", "py.PoolingRegion", json.dumps(UP_PARAMS))
network.link("temporalMemoryRegion", "unionPoolerRegion", "UniformLink", "",
srcOutput="activeCells", destInput="activeCells")
network.link("temporalMemoryRegion", "unionPoolerRegion", "UniformLink", "",
srcOutput="predictedActiveCells", destInput="predictedActiveCells")
network.initialize()
spatial = network.regions["spatialPoolerRegion"].getSelf()
# Make sure learning is enabled (this is the default)
spatial.setParameter("learningMode", 1, True)
# We want temporal anomalies so disable anomalyMode in the SP. This mode is
# used for computing anomalies in a non-temporal model.
spatial.setParameter("anomalyMode", 1, False)
# Enable topDownMode to get the predicted columns output
temporal.setParameter("topDownMode", 1, True)
# Make sure learning is enabled (this is the default)
temporal.setParameter("learningMode", 1, True)
# Enable inference mode so we get predictions
temporal.setParameter("inferenceMode", 1, True)
temporal.setParameter("computePredictedActiveCellIndices", 1, True)
union = network.regions["unionPoolerRegion"].getSelf()
# Make sure learning is enabled (this is the default)
union.setParameter("learningMode", 1, True)
return network
