def test_compileNxModel(self):
"""Check that NxModel can be compiled with Keras."""
inputShape = (7, 7, 1)
inputLayer = NxInputLayer(inputShape)
outputLayer = NxConv2D(2, 3)(inputLayer.input)
model = NxModel(inputLayer.input, outputLayer)
model.clearTemp()
def _setup_2layer_stimulus_net(inputImage,
vTh,
padding='same',
verbose=False):
"""Helper method to set up a 2-layer CNN.
The output layer has a single kernel with all weights equal to 1.
"""
assert isinstance(inputImage, np.ndarray)
inputShape = inputImage.shape
inputLayer = NxInputLayer(inputShape,
vThMant=vTh,
biasExp=0,
visualizePartitions=False)
outputLayer = NxConv2D(filters=1,
kernel_size=3,
padding=padding,
vThMant=1000,
weightExponent=0,
synapseEncoding='sparse',
kernel_initializer='ones',
bias_initializer='zeros',
visualizePartitions=False)
model = NxModel(inputLayer.input,
outputLayer(inputLayer.input),
numCandidatesToCompute=1)
mapper = model.compileModel()
if verbose:
hiddenCore = model.board.n2Chips[0].n2CoresAsList[1]
outputCore = model.board.n2Chips[0].n2CoresAsList[0]
mapper.printCore(hiddenCore, compartments=True)
mapper.printCore(outputCore, compartments=True)
outputShape = model.output_shape[1:]
# Define probes to read out current and voltages
vProbes1 = []
uProbes2 = []
for i in range(int(np.asscalar(np.prod(inputShape)))):
vProbes1.append(inputLayer[i].probe(state=ProbableStates.VOLTAGE))
for i in range(int(np.asscalar(np.prod(outputShape)))):
uProbes2.append(outputLayer[i].probe(state=ProbableStates.CURRENT))
# Set bias currents
for i, b in enumerate(np.ravel(inputImage, 'F')):
inputLayer[i].biasMant = b
inputLayer[i].phase = 2
return model, vProbes1, uProbes2
def test_synMem2(self):
"""Test issue where nxcompiler runs out of synMem.
The problem is that whenever the number of syn bits is an exact
multiple of 64, nxcompiler requires an extra mem word. E.g. 128 bits
would count as 3 words.
We have implemented a check in the NxTF compiler that allocates an
extra mem word when it detects that the numBits are an exact multiple
of 64. So this test should pass.
"""
inputLayer = NxInputLayer((1, 1, 1))
x = NxConv2D(6, (1, 1),
name='conv1',
kernel_initializer='ones',
synapseEncoding='dense1',
useSharedSign=False)(inputLayer.input)
model = NxModel(inputLayer.input, x, saveOutput=False)
model.compileModel()
model.run(1)
model.disconnect()
layers = model.partitionOptimizer.getLayers()
assert layers[1].numSynMemWords == 2
def test_synMem(self):
"""Test issue where nxcompiler runs out of synMem.
The problem is that whenever the number of syn bits is an exact
multiple of 64, nxcompiler requires an extra mem word. E.g. 128 bits
would count as 3 words.
Normally, this test would fail due to an assertion in the nxcompiler
that checks for number of syn mem words to be smaller than 2**14.
We have implemented a check in the NxTF compiler that allocates an
extra mem word when it detects that the numBits are an exact multiple
of 64. So this test should pass.
"""
np.random.seed(123)
inputLayer = NxInputLayer((1, 1, 256))
x = NxConv2D(256, (1, 1), name='conv1')(inputLayer.input)
model = NxModel(inputLayer.input, x)
model.compileModel()
model.run(1)
model.disconnect()
def test_evaluateNxModel(self):
"""Check that NxModel can be evaluated like a Keras Model."""
batchSize = 10
batchInputShape = (batchSize, 7, 7, 1)
inputLayer = NxInputLayer(batch_input_shape=batchInputShape)
layer = NxConv2D(2, 3)(inputLayer.input)
model = NxModel(inputLayer.input, layer)
model.compile('sgd', 'mse')
model.evaluate(np.random.random_sample(model.input_shape),
np.random.random_sample(model.output_shape))
model.clearTemp()
def create_snn_model():
"""
Create a Spiking Model of the MNIST dnn
:return: Spiking Model
"""
vth_mant = 2**9
bias_exp = 6
weight_exponent = 0
synapse_encoding = 'sparse'
inputLayer = NxInputLayer(input_shape, vThMant=vth_mant, biasExp=bias_exp)
layer = NxConv2D(filters=16,
kernel_size=(5, 5),
strides=(2, 2),
input_shape=input_shape,
vThMant=vth_mant,
weightExponent=weight_exponent,
synapseEncoding=synapse_encoding)(inputLayer.input)
layer = NxConv2D(filters=32,
kernel_size=(3, 3),
vThMant=vth_mant,
weightExponent=weight_exponent,
synapseEncoding=synapse_encoding)(layer)
layer = NxConv2D(filters=64,
kernel_size=(3, 3),
strides=(2, 2),
vThMant=vth_mant,
weightExponent=weight_exponent,
synapseEncoding=synapse_encoding)(layer)
layer = NxConv2D(filters=10,
kernel_size=(4, 4),
activation='softmax',
vThMant=vth_mant,
weightExponent=weight_exponent,
synapseEncoding=synapse_encoding)(layer)
spiking_model = NxModel(inputLayer.input, layer, numCandidatesToCompute=1)
spiking_model.summary()
return spiking_model
def test_compile(self):
"""Check compilation of NxModel and retrieval of cxResourceMap."""
# Create an arbitrary model
inputShape = (31, 35, 2)
inputLayer = NxInputLayer(inputShape)
hiddenLayer = NxConv2D(4, 3)
outputLayer = NxConv2D(7, 3)
model = NxModel(inputLayer.input,
outputLayer(hiddenLayer(inputLayer.input)),
numCandidatesToCompute=1)
model.compileModel()
if self.verbose:
hiddenLayer.exclusionCriteria.print()
outputLayer.exclusionCriteria.print()
for l in model.layers:
if isinstance(l, NxConv2D):
print(l._cxResourceMap)
def test_partitionModel3(self):
"""Test partitioning of a NxModel.
After completion of the algorithm, the partitioner reconstructs the
kernelIdMap from the synapses and axons generated during partitioning.
An exception is thrown if the reconstructed map does not equal the
original map.
"""
inputShape = (73, 81, 3)
inputLayer = NxInputLayer(inputShape)
hiddenLayer = NxConv2D(11,
3,
strides=(2, 2),
padding='same',
validatePartitions=True)(inputLayer.input)
hiddenLayer = NxAveragePooling2D(4,
validatePartitions=True)(hiddenLayer)
hiddenLayer = NxFlatten()(hiddenLayer)
outputLayer = NxDense(50, validatePartitions=True)(hiddenLayer)
model = NxModel(inputLayer.input, outputLayer)
model.partition()
model.clearTemp()
def test_CompartmentInterface_probe(self):
"""Check probe generation."""
# Create an arbitrary model
inputShape = (16, 16, 1)
inputLayer = NxInputLayer(inputShape)
outputLayer = NxConv2D(1, 3, padding='same', validatePartitions=True)
model = NxModel(inputLayer.input,
outputLayer(inputLayer.input),
numCandidatesToCompute=3)
model.compileModel()
uProbe = outputLayer[0].probe(ProbableStates.CURRENT)
sProbe = outputLayer[0].probe(ProbableStates.SPIKE)
aProbe = outputLayer[0].probe(ProbableStates.ACTIVITY)
pProbe = outputLayer[2].probe(ProbableStates.PHASE)
self.assertTrue(isinstance(uProbe, N2Probe))
self.assertTrue(isinstance(sProbe, N2SpikeProbe))
self.assertTrue(isinstance(aProbe, N2Probe))
self.assertTrue(isinstance(pProbe, N2Probe))
self.assertEqual(uProbe.chipId, outputLayer._cxResourceMap[0, 0])
self.assertEqual(uProbe.coreId, outputLayer._cxResourceMap[0, 1])
self.assertEqual(uProbe.nodeId, outputLayer._cxResourceMap[0, 2])
self.assertEqual(sProbe.chipId, outputLayer._cxResourceMap[0, 0])
self.assertEqual(sProbe.coreId, outputLayer._cxResourceMap[0, 1])
self.assertEqual(sProbe.cxId, outputLayer._cxResourceMap[0, 2])
self.assertEqual(aProbe.chipId, outputLayer._cxResourceMap[0, 0])
self.assertEqual(aProbe.coreId, outputLayer._cxResourceMap[0, 1])
self.assertEqual(aProbe.nodeId, outputLayer._cxResourceMap[0, 2])
self.assertEqual(pProbe.chipId, outputLayer._cxResourceMap[2, 0])
self.assertEqual(pProbe.coreId, outputLayer._cxResourceMap[2, 1])
self.assertEqual(pProbe.nodeId, outputLayer._cxResourceMap[2, 2] // 4)
def test_setter_getter(self):
"""Check setter and getter methods of CompartmentInterface."""
# Create an arbitrary model
inputShape = (16, 16, 1)
inputLayer = NxInputLayer(inputShape)
outputLayer = NxConv2D(1, 3, padding='same', validatePartitions=True)
model = NxModel(inputLayer.input,
outputLayer(inputLayer.input),
numCandidatesToCompute=3)
model.compileModel()
# Set ome arbitrary values
outputLayer[0].current = 1
outputLayer[0].voltage = 2
outputLayer[0].activity = 3
outputLayer[0].biasMant = 4
outputLayer[0].biasExp = 5
outputLayer[0].phase = 1
outputLayer[1].phase = 2
outputLayer[2].phase = 3
outputLayer[3].phase = 4
outputLayer[4].phase = 5
# Check values
self.assertEqual(outputLayer[0].current, 1)
self.assertEqual(outputLayer[0].voltage, 2)
self.assertEqual(outputLayer[0].activity, 3)
self.assertEqual(outputLayer[0].biasMant, 4)
self.assertEqual(outputLayer[0].biasExp, 5)
self.assertEqual(outputLayer[0].phase, 1)
self.assertEqual(outputLayer[1].phase, 2)
self.assertEqual(outputLayer[2].phase, 3)
self.assertEqual(outputLayer[3].phase, 4)
self.assertEqual(outputLayer[4].phase, 5)
def test_partitionPooling(self):
"""Test partitioning a single pooling layer."""
inputShape = (5, 5, 4)
inputLayer = NxInputLayer(inputShape)
outputLayer = NxAveragePooling2D(2, validatePartitions=True)
model = NxModel(inputLayer.input, outputLayer(inputLayer.input))
model.partition()
model.clearTemp()
def test_partition1(self):
"""Test partitioning a single NxConv1D layer."""
inputShape = (5, 4)
inputLayer = NxInputLayer(inputShape)
outputLayer = NxConv1D(2, 3, validatePartitions=True)
model = NxModel(inputLayer.input, outputLayer(inputLayer.input))
model.partition()
model.clearTemp()
def test_partition1(self):
"""Test partitioning a single fully-connected layer."""
inputShape = (3, 3, 2)
inputLayer = NxInputLayer(inputShape)
flattenLayer = NxFlatten()(inputLayer.input)
outputLayer = NxDense(10, validatePartitions=True)
model = NxModel(inputLayer.input, outputLayer(flattenLayer))
model.partition()
model.clearTemp()
def test_partition2(self):
"""Test partitioning two fully-connected layers."""
inputShape = (40, 32, 2)
inputLayer = NxInputLayer(inputShape)
flattenLayer = NxFlatten()(inputLayer.input)
hiddenLayer = NxDense(100, validatePartitions=True)
outputLayer = NxDense(10, validatePartitions=True)
model = NxModel(inputLayer.input,
outputLayer(hiddenLayer(flattenLayer)))
model.partition()
model.clearTemp()
def setUpDNN(self) -> ComposableDNN:
"""Sets up a DNN"""
# Specify input shape of network.
inputShape = (16, 16, 3)
#################
# BUILD NETWORK #
#################
inputLayer = NxInputLayer(inputShape)
x = NxConv2D(4, (3, 3))(inputLayer.input)
x = NxAveragePooling2D()(x)
x = NxFlatten()(x)
x = NxDense(10)(x)
DNNModel = NxModel(inputLayer.input, x)
composableDNNModel = ComposableDNN(model=DNNModel, num_steps_per_img=100)
return composableDNNModel
def test_partition2(self):
"""Test partitioning two NxConv1D layers."""
inputShape = (500, 4)
inputLayer = NxInputLayer(inputShape)
hiddenLayer = NxConv1D(20,
3,
strides=2,
padding='same',
validatePartitions=True)
outputLayer = NxConv1D(10, 2, padding='same', validatePartitions=True)
model = NxModel(inputLayer.input,
outputLayer(hiddenLayer(inputLayer.input)))
model.partition()
model.clearTemp()
def test_partitionPooling2(self):
"""Test partitioning two pooling layers."""
inputShape = (30, 40, 4)
inputLayer = NxInputLayer(inputShape)
hiddenLayer = NxAveragePooling2D(2,
padding='same',
validatePartitions=True)
outputLayer = NxAveragePooling2D(3,
strides=(1, 1),
validatePartitions=True)
model = NxModel(inputLayer.input,
outputLayer(hiddenLayer(inputLayer.input)))
model.partition()
model.clearTemp()
def test_partition2(self):
"""Test partitioning two DepthwiseConv2D layers."""
inputShape = (50, 20, 4)
inputLayer = NxInputLayer(inputShape)
hiddenLayer = NxDepthwiseConv2D(3,
strides=(2, 2),
padding='same',
validatePartitions=True)
outputLayer = NxDepthwiseConv2D(2,
padding='same',
validatePartitions=True)
model = NxModel(inputLayer.input,
outputLayer(hiddenLayer(inputLayer.input)))
model.partition()
model.clearTemp()
def test_partitionModel2(self):
"""Test partitioning of a NxModel.
After completion of the algorithm, the partitioner reconstructs the
kernelIdMap from the synapses and axons generated during partitioning.
An exception is thrown if the reconstructed map does not equal the
original map.
"""
inputShape = (2, 2, 128)
inputLayer = NxInputLayer(inputShape)
hiddenLayer = NxConv2D(128, 2, padding='same', validatePartitions=True)
hiddenLayer.exclusionCriteria.maxNumCompartments /= 4
outputLayer = NxConv2D(256, 3, padding='same', validatePartitions=True)
outputLayer.exclusionCriteria.maxNumCompartments /= 4
model = NxModel(inputLayer.input,
outputLayer(hiddenLayer(inputLayer.input)))
model.partition()
model.clearTemp()
def test_partitionModel4(self):
"""Test partitioning of a NxModel.
After completion of the algorithm, the partitioner reconstructs the
kernelIdMap from the synapses and axons generated during partitioning.
An exception is thrown if the reconstructed map does not equal the
original map.
"""
inputShape = (15, 15, 3)
inputLayer = NxInputLayer(inputShape)
outputLayer = NxConv2D(3,
3,
strides=(2, 2),
padding='same',
validatePartitions=True)(inputLayer.input)
model = NxModel(inputLayer.input,
outputLayer,
numCandidatesToCompute=10)
model.partition()
model.clearTemp()
def test_Conv2DBiases(self):
"""Test correlation between ANN activations and SNN spikerates.
The network consists of an input layer connected to a convolution
layer. The input to the network is a square gray-scale image of random
integers. The weights are also random integers. Biases are non-zero.
This test asserts that the spikerates in the output layer are close to
the ANN activations, by computing the Pearson correlation coefficient.
A perfect correlation cannot be expected due to quantization errors
when approximating ANN activations with discrete spikes. However,
correlations should be higher than 0.99.
"""
def bias_initializer(shape, dtype=None, biasScale=1):
"""Increasing integer bias initializer for Keras layer.
:param list | tuple | np.ndarray shape: Shape of biases.
:param str | type | None dtype: Data type of biases.
:param int biasScale: Scale factor applied to the biases.
:return: Bias tensor.
"""
return k.constant(np.arange(biasScale), dtype, shape)
numFilters = 16
kernelShape = (3, 3)
inputShape = (28, 28, 3)
vThMant = numFilters * 2**4
thrGain = 2**6
visualizePartitions = False
plotUV = False
numSteps = 500
bias_init = partial(bias_initializer, biasScale=numFilters)
layer = NxConv2D(filters=numFilters,
kernel_size=kernelShape,
vThMant=vThMant,
kernel_initializer='zeros',
bias_initializer=bias_init,
validatePartitions=True,
probeSpikes=True,
activation='relu',
strides=(2, 2))
inputLayer = NxInputLayer(batch_input_shape=(1, ) + inputShape,
vThMant=vThMant,
visualizePartitions=visualizePartitions)
model = NxModel(inputLayer.input, layer(inputLayer.input))
model.compileModel()
outputShape = layer.output_shape[1:]
# Define probes to read out currents.
vProbes = []
sProbes = []
for i in range(int(np.asscalar(np.prod(outputShape)))):
vProbes.append(layer[i].probe(ProbableStates.VOLTAGE))
sProbes.append(layer[i].probe(ProbableStates.ACTIVITY))
model.run(numSteps)
model.disconnect()
data = extract(sProbes)
spikecount = _data_to_img(data // 127, outputShape)
spikerates = spikecount / numSteps
batchInputImage = np.expand_dims(np.zeros(inputShape), 0)
activations = model.predict(batchInputImage)[0] / (vThMant * thrGain)
if plotUV:
plt.figure(1)
_plot_stimulus_response(vProbes, sProbes)
plt.show()
plt.figure(2)
plt.plot(activations.flatten(), spikerates.flatten(), '.')
plt.show()
plt.figure(3)
plt.imshow(normalize_image_dims(activations))
plt.show()
plt.figure(4)
plt.imshow(normalize_image_dims(spikerates))
plt.show()
corr = np.corrcoef(np.ravel(spikerates), np.ravel(activations))[0, 1]
self.assertAlmostEqual(corr,
1,
2,
msg="Correlation between ANN activations "
"and SNN spikerates is too low.")
def test_saveLoadNxModel(self):
"""Check that NxModel can be saved and loaded like a Keras Model."""
inputLayer = NxInputLayer(batch_input_shape=(1, 10, 10, 3))
layer = NxConv2D(2, 3)(inputLayer.input)
model1 = NxModel(inputLayer.input, layer)
model1.compile('sgd', 'mse')
model1.compileModel()
model1.clearTemp()
filename = os.path.abspath(
os.path.join(os.path.dirname(os.path.realpath(__file__)),
'../../..', 'temp', str(hash(model1))))
model1.save(filename)
model2 = loadNxModel(filename)
os.remove(filename)
x = np.random.random_sample(model1.input_shape)
y1 = model1.predict(x)
y2 = model2.predict(x)
self.assertTrue(np.array_equal(y1, y2))
def test_Flatten(self):
"""Test correlation between ANN activations and SNN spikerates.
The network consists of a 3D input layer, followed by a flatten layer,
which has 1-to-1 connections to the output layer. The input pixel
values are set to ascending integers in Fortran style.
This test asserts that the spikerates in the output layer are close to
the ANN activations, by computing the Pearson correlation coefficient.
A perfect correlation cannot be expected due to quantization errors
when approximating ANN activations with discrete spikes. However,
correlations should be higher than 0.99.
"""
visualizePartitions = False
doPlot = False
# Height, width, depth
inputShape = (3, 4, 5)
numInputNeurons = int(np.asscalar(np.prod(inputShape)))
numOutputNeurons = numInputNeurons - 1
inputScale = 255
thrToInputRatio = 2**7
thrGain = 2**6
# No need to divide by thrGain because spike input receives equal gain.
vThMant = 1
vThMantInput = thrToInputRatio * inputScale // thrGain
maxNumSpikes = 100
numSteps = thrToInputRatio * maxNumSpikes
weights = np.eye(numInputNeurons, numOutputNeurons, dtype=int)
biases = np.zeros(numOutputNeurons, int)
nxInput = NxInputLayer(inputShape,
vThMant=vThMantInput,
visualizePartitions=visualizePartitions)
nxLayer = NxDense(numOutputNeurons,
weights=[weights, biases],
vThMant=vThMant,
validatePartitions=True,
probeSpikes=True)
nxModel = NxModel(nxInput.input, nxLayer(NxFlatten()(nxInput.input)))
nxModel.compileModel()
kerasInput = Input(inputShape)
kerasLayer = Dense(numOutputNeurons,
weights=[weights, biases])(Flatten()(kerasInput))
kerasModel = Model(kerasInput, kerasLayer)
# Define probes to read out currents.
sProbes = []
for i in range(numOutputNeurons):
sProbes.append(nxLayer[i].probe(ProbableStates.ACTIVITY))
# Set bias currents
inputImage = np.reshape(np.arange(numInputNeurons), inputShape, 'F')
inputImage = inputImage % 255
for i, b in enumerate(np.ravel(inputImage, 'F')):
nxInput[i].biasMant = b
nxInput[i].phase = 2
nxModel.run(numSteps)
nxModel.disconnect()
data = extract(sProbes)
spikecount = _data_to_img(data // 127, nxLayer.output_shape[1:])
spikerates = spikecount / numSteps * thrToInputRatio
batchInputImage = np.expand_dims(inputImage, 0)
activations = \
kerasModel.predict(batchInputImage)[0] / (vThMant * thrGain)
if doPlot:
plt.figure(3)
plt.imshow(normalize_image_dims(inputImage))
plt.show()
plt.figure(6)
plt.plot(activations.flatten(), spikerates.flatten(), '.')
plt.show()
plt.figure(7)
plt.imshow(normalize_image_dims(activations))
plt.show()
plt.figure(8)
plt.imshow(normalize_image_dims(spikerates))
plt.show()
corr = np.corrcoef(np.ravel(spikerates), np.ravel(activations))[0, 1]
self.assertAlmostEqual(corr,
1,
2,
msg="Correlation between ANN activations "
"and SNN spikerates is too low.")
def test_random_cor(self):
"""
Tests the soft reset mode by comparing activations from an ANN
to spike-rate from the converted SNN. The input and weights
are randomly initialized.
"""
seed = 123
np.random.seed(seed)
plot = False
verbose = False
visualizePartitions = False
logger = None
resetMode = 'soft'
neuronSize = 2 if resetMode == 'soft' else 1
inputShape = (4, 4, 1)
inputScale = 255
inputImage = np.random.randint(int(inputScale * 0.25), int(inputScale),
inputShape)
maxNumSpikes = 100
numSteps = int(np.max(inputImage / 255) * maxNumSpikes)
inputLayer = NxInputLayer(batch_input_shape=(1, ) + inputShape,
vThMant=255,
visualizePartitions=visualizePartitions,
resetMode=resetMode,
probeSpikes=True)
out = inputLayer.input
layers = [inputLayer]
# Conv2D
kernelShape = (3, 3, 1)
kernelScale = 4
# No need to divide by thrGain because spike input receives equal gain.
vThMant = 2**9 - 1
kernel_init = partial(kernel_initializer, kernelScale=kernelScale)
numLayers = 1
for i in range(numLayers):
layer = NxConv2D(filters=kernelShape[-1],
kernel_size=kernelShape[:-1],
vThMant=vThMant,
kernel_initializer=kernel_init,
bias_initializer='ones',
validatePartitions=False,
probeSpikes=True,
activation='relu',
resetMode=resetMode)
layers.append(layer)
out = layer(out)
model = NxModel(inputLayer.input, out, logger=logger)
for layer in layers[1:]:
weights, biases = layer.get_weights()
weights, biases = to_integer(weights, biases, 8,
np.max(weights) // 2)
layer.set_weights([weights, biases])
mapper = model.compileModel()
if verbose:
printLayerMappings(layers, mapper, synapses=True, inputAxons=True)
printLayers(layers)
print(model.summary())
layerProbes = []
for layer in layers:
shape = layer.output_shape[1:]
# Define probes to read out currents.
vProbes = []
sProbes = []
uProbes = []
for i in range(int(np.asscalar(np.prod(shape))) * neuronSize):
vProbes.append(layer[i].probe(ProbableStates.VOLTAGE))
sProbes.append(layer[i].probe(ProbableStates.ACTIVITY))
uProbes.append(layer[i].probe(ProbableStates.CURRENT))
layerProbes.append([uProbes, vProbes, sProbes])
# Set bias currents
for i, b in enumerate(np.ravel(inputImage, 'F')):
inputLayer[i * neuronSize].biasMant = b
inputLayer[i * neuronSize].phase = 2
if verbose:
for layer in layers:
print(getCompartmentStates(layer, neuronSize))
model.run(numSteps)
if verbose:
for layer in layers:
print(getCompartmentStates(layer, neuronSize))
model.disconnect()
data = [[extract(probe) for probe in lp] for lp in layerProbes]
if plot:
plotLayerProbes(layers, data, neuronSize)
spikesRates = []
for i, (layer, d) in enumerate(zip(layers, data)):
sData = d[2][:, (neuronSize - 1)::neuronSize]
shape = layer.output_shape[1:]
spikecount = _data_to_img(sData // 127, shape)
spikesRates.append(spikecount / numSteps)
batchInputImage = np.expand_dims(inputImage, 0)
activations = model.predict(batchInputImage)[0]
if plot:
plt.figure()
plt.plot(inputImage.flatten(), spikesRates[0].flatten(), '.')
plt.show()
plt.figure()
plt.plot(activations.flatten(), spikesRates[-1].flatten(), '.')
plt.show()
plt.figure()
plt.imshow(normalize_image_dims(activations))
plt.show()
plt.figure()
plt.imshow(normalize_image_dims(spikesRates[-1]))
plt.show()
cor = np.corrcoef(np.ravel(spikesRates[-1]), np.ravel(activations))[0,
1]
self.assertGreater(cor, 0.99)
if verbose:
print(cor)
weights = weights.transpose()
num_neurons = num_classes
biases = np.zeros(num_neurons)
layer = NxDense(num_neurons,
weights=[weights, biases],
compartmentKwargs=compartment_kwargs,
connectionKwargs=conn_kwargs,
resetMode=reset_mode,
name=name)(layer)
# COMPILATION #
###############
# Compile complete model.
nxmodel = NxModel(input_layer.input, layer)
nxmodel.summary()
nxmodel.compile() # Keras compile function. Not yet invoking NxTF compiler.
# Apply pooling weights.
for name, weights in pool_weights.items():
nxmodel.get_layer(name).set_weights(weights)
# Use the composablility interface of NxSDK to wrap our NxTF model. Allows
# more efficient input injection, output readout, and resetting via snips.
nxmodel_composable = ComposableDNN(nxmodel,
num_steps_per_img,
enable_reset=reset_between_samples)
model = ComposableModel('gestures_cnn')
input_generator = SpikeInputGenerator(name='SpikeGen',
packetSize=256,
conn_kwargs['signMode'] = sign_mode
weights = weights.transpose()
num_neurons = num_classes
layer = NxDense(num_neurons,
weights=[weights, np.zeros(num_neurons)],
compartmentKwargs=compartment_kwargs,
connectionKwargs=conn_kwargs,
resetMode=reset_mode,
name=name)(layer)
# COMPILATION #
###############
# Compile complete model.
nxmodel = NxModel(input_layer.input, layer)
nxmodel.summary()
nxmodel.compile() # Keras compile function. Not yet invoking NxTF compiler.
# Use the composablility interface of NxSDK to wrap our NxTF model. Allows
# more efficient input injection, output readout, and resetting via snips.
nxmodel_composable = ComposableDNN(nxmodel,
num_steps_per_img,
enable_reset=reset_between_samples)
model = ComposableModel('nmnist_mlp')
input_generator = SpikeInputGenerator(name='SpikeGen',
packetSize=256,
numSnipsPerChip=3,
queueSize=512)
model.add(nxmodel_composable)
def test_NxInputLayer(self):
"""
Test input layer in soft-reset mode.
"""
plot = False
verbose = False
resetMode = 'soft'
neuronSize = 2 if resetMode == 'soft' else 1
input_shape = (32, 32, 3)
inputSize = np.prod(input_shape)
inputLayer = NxInputLayer(input_shape,
probeSpikes=True,
vThMant=255,
resetMode=resetMode)
out = inputLayer.input
model = NxModel(out, out)
model.compile('adam', 'categorical_crossentropy', ['accuracy'])
model.summary()
layers = [inputLayer]
input_image = np.linspace(
0, 256, endpoint=False,
num=inputSize).reshape(input_shape).astype(int)
x_test = [input_image]
y_test = [0]
mapper = model.compileModel()
if verbose:
printLayerMappings(layers, mapper, synapses=True, inputAxons=True)
printLayers(layers)
layerProbes = []
numProbes = 32
for i, layer in enumerate(layers):
shape = layer.output_shape[1:]
# Define probes to read out currents.
vProbes = []
sProbes = []
uProbes = []
toProbe = numProbes if i == (len(layers) -
1) else numProbes * neuronSize
toProbe = min(toProbe,
int(np.asscalar(np.prod(shape))) * neuronSize)
for j in range(toProbe):
vProbes.append(layer[j].probe(ProbableStates.VOLTAGE))
sProbes.append(layer[j].probe(ProbableStates.ACTIVITY))
uProbes.append(layer[j].probe(ProbableStates.CURRENT))
layerProbes.append([uProbes, vProbes, sProbes])
# How many time steps to run each sample.
num_steps = 1000
# How many samples to test.
num_samples_to_test = 1
# Iterate over samples in testset.
for i, (input_image, target) in enumerate(zip(x_test, y_test)):
if i == num_samples_to_test:
break
if plot:
plt.hist(input_image.ravel())
plt.show()
# Set input bias currents.
for j, b in enumerate(np.ravel(input_image, 'F')):
inputLayer[j * neuronSize].biasMant = b
inputLayer[j * neuronSize].biasExp = 6
inputLayer[j * neuronSize].phase = 2
# Run model.
model.run(num_steps)
# Clean up.
data = [[extract(probe) for probe in lp] for lp in layerProbes]
spikesRates = []
for i, (layer, d) in enumerate(zip(layers, data)):
sData = d[2][:, (neuronSize - 1)::neuronSize]
spikecount = (sData // 127).sum(0)
spikesRates.append(spikecount / num_steps)
model.disconnect()
layer_activations = [x_test[0]]
if plot:
for activations, spikerate in zip(layer_activations, spikesRates):
plt.figure()
scale = np.max(activations)
spikesFlat = spikerate #spikerate.flatten()
plt.plot(
activations.flatten('F')[:len(spikesFlat)] / scale,
spikesFlat, '.')
plotLayerProbes(layers, data, neuronSize)
cor = np.corrcoef(np.ravel(spikesRates[-1]),
np.ravel(layer_activations[0],
'F')[:numProbes // 2])[0, 1]
self.assertGreater(cor, 0.99)
def runCorrelationRandom(layer,
vThMant,
insertFlatten=False,
inputShape=None,
logger=None):
"""Run network to test correlation between ANN and SNN.
:param NxLayer | Layer layer: NxLayer to test.
:param int vThMant: Threshold of ``layer``; used to scale activations.
:param bool insertFlatten: Whether to flatten input before applying it to
``layer``.
:param np.ndarray | tuple | list inputShape: Shape of input to the network.
:param logging.Logger logger: Logger.
:return: Pearson correlation coefficient of ANN activations and SNN rates.
:rtype: float
"""
seed = 123
np.random.seed(seed)
visualizePartitions = False
plotUV = False
if inputShape is None:
inputShape = (7, 7, 1)
numInputNeurons = int(np.asscalar(np.prod(inputShape)))
inputScale = numInputNeurons - 1
thrToInputRatio = 2**7
thrGain = 2**0
vThMantInput = thrToInputRatio * inputScale // thrGain
maxNumSpikes = 100
numSteps = thrToInputRatio * maxNumSpikes
inputImage = np.random.randint(0, inputScale, inputShape)
inputLayer = NxInputLayer(batch_input_shape=(1, ) + inputShape,
vThMant=vThMantInput,
visualizePartitions=visualizePartitions)
out = layer(NxFlatten()(inputLayer.input)) \
if insertFlatten else layer(inputLayer.input)
model = NxModel(inputLayer.input, out, logger=logger)
model.compileModel()
outputShape = layer.output_shape[1:]
# Define probes to read out currents.
vProbes0 = []
for i in range(numInputNeurons):
vProbes0.append(inputLayer[i].probe(ProbableStates.VOLTAGE))
vProbes = []
sProbes = []
for i in range(int(np.asscalar(np.prod(outputShape)))):
vProbes.append(layer[i].probe(ProbableStates.VOLTAGE))
sProbes.append(layer[i].probe(ProbableStates.ACTIVITY))
# Set bias currents
for i, b in enumerate(np.ravel(inputImage, 'F')):
inputLayer[i].biasMant = b
inputLayer[i].phase = 2
model.run(numSteps)
model.disconnect()
data = extract(sProbes)
spikecount = _data_to_img(data // 127, outputShape)
spikerates = spikecount / numSteps * thrToInputRatio
batchInputImage = np.expand_dims(inputImage, 0)
activations = model.predict(batchInputImage)[0] / (vThMant * thrGain)
if plotUV:
plt.figure(1)
_plot_stimulus_response(vProbes0, [])
plt.show()
plt.figure(2)
_plot_stimulus_response(vProbes, sProbes)
plt.show()
plt.figure(3)
plt.imshow(normalize_image_dims(inputImage))
plt.show()
plt.figure(6)
plt.plot(activations.flatten(), spikerates.flatten(), '.')
plt.show()
plt.figure(7)
plt.imshow(normalize_image_dims(activations))
plt.show()
plt.figure(8)
plt.imshow(normalize_image_dims(spikerates))
plt.show()
return np.corrcoef(np.ravel(spikerates), np.ravel(activations))[0, 1]
'functionalState': 0,
'enableSomaTrace': True,
'refractoryDelay': 0,
}
num_input_neurons = 1024
num_output_neurons = 1
input_layer = NxInputLayer(num_input_neurons,
compartmentKwargs=compartment_kwargs,
inputMode=InputModes.AEDAT)
layer = NxDense(num_output_neurons,
compartmentKwargs=compartment_kwargs)(input_layer.input)
nxmodel = NxModel(input_layer.input, layer)
nxmodel.summary()
nxmodel.compile()
num_steps_per_img = num_input_neurons + 1
nxmodel_composable = ComposableDNN(nxmodel, num_steps_per_img)
input_generator = SpikeInputGenerator(name='SpikeGen',
packetSize=256,
queueSize=64,
numSnipsPerChip=1)
input_generator.connect(nxmodel_composable)
input_generator.processes.inputEncoder.executeAfter(
nxmodel_composable.processes.reset)
def runModelFromConfig(argsInput, argsHidden, argsOutput, n=None):
"""A helper function used by test_correlationUniform.
Given a model configuration, this function creates a NxModel, checks
that the model architecture is valid, then partitions, maps and runs the
network with a constant white input image.
The function probes voltage and current states, which can be plotted.
Spikecountss are derived from the current trace.and are used to assert
that the SNN activity equals the ANN activations.
:param InputParams argsInput: Parameters for input layer.
:param LayerParams argsHidden: Parameters for hidden layer.
:param LayerParams argsOutput: Parameters for output layer.
:param int n: The id of the current config. Only used for printing.
"""
# Define stimulus and runtime settings.
# #####################################
inputScale = 1
vThMantInput = 1
vThMant = 1
numSteps = (vThMantInput << 6) // inputScale + 1 + 1 + 1 + 1
biasExp = 0
visualizePartitions = False
plotUV = False
# Build model.
# ############
inputShape = (argsInput.imgHeight, argsInput.imgWidth, argsInput.imgDepth)
batchInputShape = (1, ) + inputShape
inputLayer = NxInputLayer(batch_input_shape=batchInputShape,
biasExp=biasExp,
vThMant=vThMantInput,
visualizePartitions=visualizePartitions)
inputLayer._maxNumCompartments = argsInput.maxNumCxPerCore
hiddenLayer = NxConv2D(argsHidden.kernelDepth,
(argsHidden.kernelHeight, argsHidden.kernelWidth),
strides=(argsHidden.strideY, argsHidden.strideX),
padding=argsHidden.padding,
vThMant=vThMant,
synapseEncoding=argsHidden.encoding,
kernel_initializer='ones',
visualizePartitions=visualizePartitions,
validatePartitions=True)
hiddenLayer._maxNumCompartments = argsHidden.maxNumCxPerCore
outputLayer = NxConv2D(argsOutput.kernelDepth,
(argsOutput.kernelHeight, argsOutput.kernelWidth),
strides=(argsOutput.strideY, argsOutput.strideX),
padding=argsOutput.padding,
vThMant=vThMant,
synapseEncoding=argsOutput.encoding,
kernel_initializer='ones',
validatePartitions=True)
outputLayer._maxNumCompartments = argsOutput.maxNumCxPerCore
hiddenShape = hiddenLayer.compute_output_shape(batchInputShape)
outputShape = outputLayer.compute_output_shape(hiddenShape)
hiddenShape = hiddenShape[1:]
outputShape = outputShape[1:]
if np.any(np.array(hiddenShape) <= 0) or \
np.any(np.array(outputShape) <= 0):
print("Configuration resulted in negative layer shape; skip.")
print("Input layer shape: {}".format(inputShape))
print("Hidden layer shape: {}".format(hiddenShape))
print("Output layer shape: {}".format(outputShape))
return
# Create NxModel from input to output layer.
snn = NxModel(inputLayer.input,
outputLayer(hiddenLayer(inputLayer.input)),
numCandidatesToCompute=1)
snn.compileModel()
# Create plain Keras model from input to hidden layer so we
# can read out hidden layer activations.
model1 = Model(inputLayer.input, hiddenLayer(inputLayer.input))
hiddenInput = Input(hiddenShape)
model2 = Model(hiddenInput, outputLayer(hiddenInput))
# Define probes to read out currents.
# ###################################
u = ProbableStates.CURRENT
v = ProbableStates.VOLTAGE
uProbes1 = []
vProbes1 = []
for i in range(int(np.asscalar(np.prod(hiddenShape)))):
uProbes1.append(hiddenLayer[i].probe(u))
if plotUV:
vProbes1.append(hiddenLayer[i].probe(v))
uProbes2 = []
vProbes2 = []
for i in range(int(np.asscalar(np.prod(outputShape)))):
uProbes2.append(outputLayer[i].probe(u))
if plotUV:
vProbes2.append(outputLayer[i].probe(v))
# Apply input image via bias currents.
# ####################################
inputImage = inputScale * np.ones(inputShape, int)
for i, biasMant in enumerate(np.ravel(inputImage, 'F')):
inputLayer[i].biasMant = biasMant
inputLayer[i].phase = 2
# Run model.
# ##########
snn.run(numSteps)
snn.disconnect()
# Get SNN spike counts.
data1 = extract(uProbes1)
spikecount1 = _data_to_img(data1 // (vThMant * 2**6), hiddenShape)
data2 = extract(uProbes2)
spikecount2 = _data_to_img(data2 // (vThMant * 2**6), outputShape)
# Get ANN activations as target reference.
batchInputImage = np.expand_dims(inputImage, 0)
activations1 = model1.predict(batchInputImage)
activations2 = model2.predict(activations1 > 0)
# Remove batch dim.
activations1 = activations1[0].astype(int)
activations2 = activations2[0].astype(int)
# Plotting.
# #########
if plotUV:
plt.figure()
_plot_stimulus_response(vProbes1, uProbes1)
plt.show()
plt.figure()
_plot_stimulus_response(vProbes2, uProbes2)
plt.show()
# Validation.
# ###########
def getErrorMessage(spikecounts, activations):
"""Generate an error message for failed test.
The error message contains a code snippet to reproduce failure.
:param np.ndarray spikecounts: SNN spike counts of layer.
:param np.ndarray activations: ANN activations of layer.
:return: Error message.
:rtype: str
"""
return ("SNN spikecounts not equal to ANN activations.\n" +
"Spikecounts: \n{}\n\n".format(spikecounts) +
"Activations: \n{}\n\n".format(activations) +
"Use the following code to reproduce the error.\n\n" +
"def test_correlation{}():\n".format('' if n is None else n) +
"\targsInput = {}\n".format(argsInput) +
"\targsHidden = {}\n".format(argsHidden) +
"\targsOutput = {}\n".format(argsOutput) +
"\trunModelFromConfig(argsInput, argsHidden, argsOutput)\n\n")
assert np.array_equal(spikecount1, activations1), \
getErrorMessage(spikecount1, activations1)
assert np.array_equal(spikecount2, activations2), \
getErrorMessage(spikecount2, activations2)
