def test_add_objects(self):
network = Network(dt=1.0, learning=False)
inpt = Input(100)
network.add_layer(inpt, name="X")
lif = LIFNodes(50)
network.add_layer(lif, name="Y")
assert inpt == network.layers["X"]
assert lif == network.layers["Y"]
conn = Connection(inpt, lif)
network.add_connection(conn, source="X", target="Y")
assert conn == network.connections[("X", "Y")]
monitor = Monitor(lif, state_vars=["s", "v"])
network.add_monitor(monitor, "Y")
assert monitor == network.monitors["Y"]
network.save("net.pt")
_network = load("net.pt", learning=True)
assert _network.learning
assert "X" in _network.layers
assert "Y" in _network.layers
assert ("X", "Y") in _network.connections
assert "Y" in _network.monitors
del _network
os.remove("net.pt")
def test_add_objects(self):
network = Network(dt=1.0, learning=False)
inpt = Input(100)
network.add_layer(inpt, name='X')
lif = LIFNodes(50)
network.add_layer(lif, name='Y')
assert inpt == network.layers['X']
assert lif == network.layers['Y']
conn = Connection(inpt, lif)
network.add_connection(conn, source='X', target='Y')
assert conn == network.connections[('X', 'Y')]
monitor = Monitor(lif, state_vars=['s', 'v'])
network.add_monitor(monitor, 'Y')
assert monitor == network.monitors['Y']
network.save('net.pt')
_network = load('net.pt', learning=True)
assert _network.learning
assert 'X' in _network.layers
assert 'Y' in _network.layers
assert ('X', 'Y') in _network.connections
assert 'Y' in _network.monitors
del _network
os.remove('net.pt')
def test_add_objects(self):
network = Network(dt=1.0)
inpt = Input(100); network.add_layer(inpt, name='X')
lif = LIFNodes(50); network.add_layer(lif, name='Y')
assert inpt == network.layers['X']
assert lif == network.layers['Y']
conn = Connection(inpt, lif); network.add_connection(conn, source='X', target='Y')
assert conn == network.connections[('X', 'Y')]
monitor = Monitor(lif, state_vars=['s', 'v']); network.add_monitor(monitor, 'Y')
assert monitor == network.monitors['Y']
def main(n_input=1, n_output=10, time=1000):
# Network building.
network = Network(dt=1.0)
input_layer = RealInput(n=n_input)
output_layer = LIFNodes(n=n_output)
connection = Connection(source=input_layer, target=output_layer)
monitor = Monitor(obj=output_layer, state_vars=('v', ), time=time)
# Adding network components.
network.add_layer(input_layer, name='X')
network.add_layer(output_layer, name='Y')
network.add_connection(connection, source='X', target='Y')
network.add_monitor(monitor, name='X_monitor')
# Creating real-valued inputs and running simulation.
inpts = {'X': torch.ones(time, n_input)}
network.run(inpts=inpts, time=time)
# Plot voltage activity.
plt.plot(monitor.get('v').numpy().T)
plt.show()
def __init__(self, parameters: BenchmarkParameters):
super(BindsNetModule, self).__init__()
network = Network(batch_size=parameters.batch_size, dt=parameters.dt)
lif_nodes = LIFNodes(n=parameters.features)
monitor = Monitor(obj=lif_nodes,
state_vars=("s"),
time=parameters.sequence_length)
network.add_layer(Input(n=parameters.features), name="Input")
network.add_layer(lif_nodes, name="Neurons")
network.add_connection(
Connection(source=network.layers["Input"],
target=network.layers["Neurons"]),
source="Input",
target="Neurons",
)
network.add_monitor(monitor, "Monitor")
network.to(parameters.device)
self.parameters = parameters
self.network = network
self.monitor = monitor
n_filters * conv_size2[0] * conv_size2[1])
recurrent_conn2 = Connection(conv_layer2, conv_layer2, w=w)
network.add_layer(input_layer, name='X')
network.add_layer(conv_layer, name='Y')
network.add_layer(conv_layer_, name='Y_')
network.add_layer(conv_layer2, name='Z')
network.add_connection(conv_conn, source='X', target='Y')
network.add_connection(conv_conn_, source='X', target='Y_')
network.add_connection(recurrent_conn, source='Y', target='Y')
network.add_connection(conv_conn2, source='Y', target='Z')
network.add_connection(recurrent_conn2, source='Z', target='Z')
# Voltage recording for excitatory and inhibitory layers.
voltage_monitor = Monitor(network.layers['Y'], ['v'], time=time)
network.add_monitor(voltage_monitor, name='output_voltage')
# Load MNIST data.
dataset = MNIST(path=os.path.join('..', '..', 'data', 'MNIST'), download=True)
if train:
images, labels = dataset.get_train()
else:
images, labels = dataset.get_test()
images *= intensity
# Record spikes during the simulation.
spike_record = torch.zeros(update_interval, time, n_neurons)
# Neuron assignments and spike proportions.
network.add_connection(connection=recurrent_connection, source="B", target="B")
# Create and add input and output layer monitors.
source_monitor = Monitor(
obj=source_layer,
state_vars=("s", ), # Record spikes and voltages.
time=time, # Length of simulation (if known ahead of time).
)
target_monitor = Monitor(
obj=target_layer,
state_vars=("s", "v"), # Record spikes and voltages.
time=time, # Length of simulation (if known ahead of time).
)
network.add_monitor(monitor=source_monitor, name="A")
network.add_monitor(monitor=target_monitor, name="B")
# Create input spike data,
# where each spike is distributed according to Bernoulli(0.1).
input_data = torch.ones(time, source_layer.n).byte()
inputs = {"A": input_data}
spikes = {}
voltages = {}
# Simulate network on input data.
# network.run(inputs=inputs, time=time)
for i in range(epoch):
print(network.connections[("A", "B")].w)
network.run(inputs=inputs, time=time, input_time_dim=1)
# Retrieve and plot simulation spike, voltage data from monitors.
# voltages = {"B": target_monitor.get("v")}
output_layer = IFNodes(n=10, sum_input=True)
bias = RealInput(n=1, sum_input=True)
network.add_layer(input_layer, name='X')
network.add_layer(output_layer, name='Y')
network.add_layer(bias, name='Y_b')
input_connection = Connection(source=input_layer, target=output_layer, norm=150, wmin=-1, wmax=1)
bias_connection = Connection(source=bias, target=output_layer)
network.add_connection(input_connection, source='X', target='Y')
network.add_connection(bias_connection, source='Y_b', target='Y')
# State variable monitoring.
time = 25
for l in network.layers:
m = Monitor(network.layers[l], state_vars=['s'], time=time)
network.add_monitor(m, name=l)
# Load Fashion-MNIST data.
images, labels = FashionMNIST(path='../../data/FashionMNIST', download=True).get_train()
# Run training.
grads = {}
lr, lr_decay = 1e-2, 0.95
criterion = torch.nn.CrossEntropyLoss()
spike_ims, spike_axes, weights_im = None, None, None
for i, (image, label) in enumerate(zip(images.view(-1, 784) / 255, labels)):
label = torch.Tensor([label]).long()
# Run simulation for single datum.
inpts = {'X': image.repeat(time, 1), 'Y_b': torch.ones(time, 1)}
network.run(inpts=inpts, time=time)
def main(seed=0,
n_neurons=100,
n_train=60000,
n_test=10000,
inhib=250,
time=50,
lr=1e-2,
lr_decay=0.99,
dt=1,
theta_plus=0.05,
theta_decay=1e-7,
progress_interval=10,
update_interval=250,
train=True,
plot=False,
gpu=False):
assert n_train % update_interval == 0 and n_test % update_interval == 0, \
'No. examples must be divisible by update_interval'
params = [
seed, n_neurons, n_train, inhib, time, lr, lr_decay, theta_plus,
theta_decay, progress_interval, update_interval
]
test_params = [
seed, n_neurons, n_train, n_test, inhib, time, lr, lr_decay,
theta_plus, theta_decay, progress_interval, update_interval
]
model_name = '_'.join([str(x) for x in params])
np.random.seed(seed)
if gpu:
torch.set_default_tensor_type('torch.cuda.FloatTensor')
torch.cuda.manual_seed_all(seed)
else:
torch.manual_seed(seed)
if train:
n_examples = n_train
else:
n_examples = n_test
n_sqrt = int(np.ceil(np.sqrt(n_neurons)))
n_classes = 10
# Build network.
if train:
network = Network(dt=dt)
input_layer = Input(n=784, traces=True, trace_tc=5e-2)
network.add_layer(input_layer, name='X')
output_layer = DiehlAndCookNodes(n=n_neurons,
traces=True,
rest=0,
reset=0,
thresh=1,
refrac=0,
decay=1e-2,
trace_tc=5e-2,
theta_plus=theta_plus,
theta_decay=theta_decay)
network.add_layer(output_layer, name='Y')
w = 0.3 * torch.rand(784, n_neurons)
input_connection = Connection(source=network.layers['X'],
target=network.layers['Y'],
w=w,
update_rule=PostPre,
nu=[0, lr],
wmin=0,
wmax=1,
norm=78.4)
network.add_connection(input_connection, source='X', target='Y')
w = -inhib * (torch.ones(n_neurons, n_neurons) -
torch.diag(torch.ones(n_neurons)))
recurrent_connection = Connection(source=network.layers['Y'],
target=network.layers['Y'],
w=w,
wmin=-inhib,
wmax=0)
network.add_connection(recurrent_connection, source='Y', target='Y')
else:
path = os.path.join('..', '..', 'params', data, model)
network = load_network(os.path.join(path, model_name + '.pt'))
network.connections['X', 'Y'].update_rule = NoOp(
connection=network.connections['X', 'Y'],
nu=network.connections['X', 'Y'].nu)
network.layers['Y'].theta_decay = 0
network.layers['Y'].theta_plus = 0
# Load Fashion-MNIST data.
dataset = FashionMNIST(path=os.path.join('..', '..', 'data',
'FashionMNIST'),
download=True)
if train:
images, labels = dataset.get_train()
else:
images, labels = dataset.get_test()
images = images.view(-1, 784)
images = images / 255
# if train:
# for i in range(n_neurons):
# network.connections['X', 'Y'].w[:, i] = images[i] + images[i].mean() * torch.randn(784)
# Record spikes during the simulation.
spike_record = torch.zeros(update_interval, time, n_neurons)
# Neuron assignments and spike proportions.
if train:
assignments = -torch.ones_like(torch.Tensor(n_neurons))
proportions = torch.zeros_like(torch.Tensor(n_neurons, n_classes))
rates = torch.zeros_like(torch.Tensor(n_neurons, n_classes))
ngram_scores = {}
else:
path = os.path.join('..', '..', 'params', data, model)
path = os.path.join(path, '_'.join(['auxiliary', model_name]) + '.pt')
assignments, proportions, rates, ngram_scores = torch.load(
open(path, 'rb'))
# Sequence of accuracy estimates.
curves = {'all': [], 'proportion': [], 'ngram': []}
if train:
best_accuracy = 0
spikes = {}
for layer in set(network.layers):
spikes[layer] = Monitor(network.layers[layer],
state_vars=['s'],
time=time)
network.add_monitor(spikes[layer], name='%s_spikes' % layer)
# Train the network.
if train:
print('\nBegin training.\n')
else:
print('\nBegin test.\n')
inpt_axes = None
inpt_ims = None
spike_ims = None
spike_axes = None
weights_im = None
assigns_im = None
perf_ax = None
start = t()
for i in range(n_examples):
if i % progress_interval == 0 and train:
network.connections['X', 'Y'].update_rule.nu[1] *= lr_decay
if i % progress_interval == 0:
print(f'Progress: {i} / {n_examples} ({t() - start:.4f} seconds)')
start = t()
if i % update_interval == 0 and i > 0:
if i % len(labels) == 0:
current_labels = labels[-update_interval:]
else:
current_labels = labels[i % len(images) - update_interval:i %
len(images)]
# Update and print accuracy evaluations.
curves, predictions = update_curves(curves,
current_labels,
n_classes,
spike_record=spike_record,
assignments=assignments,
proportions=proportions,
ngram_scores=ngram_scores,
n=2)
print_results(curves)
if train:
if any([x[-1] > best_accuracy for x in curves.values()]):
print(
'New best accuracy! Saving network parameters to disk.'
)
# Save network to disk.
path = os.path.join('..', '..', 'params', data, model)
if not os.path.isdir(path):
os.makedirs(path)
network.save(os.path.join(path, model_name + '.pt'))
path = os.path.join(
path, '_'.join(['auxiliary', model_name]) + '.pt')
torch.save((assignments, proportions, rates, ngram_scores),
open(path, 'wb'))
best_accuracy = max([x[-1] for x in curves.values()])
# Assign labels to excitatory layer neurons.
assignments, proportions, rates = assign_labels(
spike_record, current_labels, n_classes, rates)
# Compute ngram scores.
ngram_scores = update_ngram_scores(spike_record,
current_labels, n_classes,
2, ngram_scores)
print()
# Get next input sample.
image = images[i % n_examples]
sample = rank_order(datum=image, time=time, dt=dt)
inpts = {'X': sample}
# Run the network on the input.
network.run(inpts=inpts, time=time)
retries = 0
while spikes['Y'].get('s').sum() < 5 and retries < 3:
retries += 1
image *= 2
sample = rank_order(datum=image, time=time, dt=dt)
inpts = {'X': sample}
network.run(inpts=inpts, time=time)
# Add to spikes recording.
spike_record[i % update_interval] = spikes['Y'].get('s').t()
# Optionally plot various simulation information.
if plot:
_input = images[i % n_examples].view(28, 28)
reconstruction = inpts['X'].view(time, 784).sum(0).view(28, 28)
_spikes = {layer: spikes[layer].get('s') for layer in spikes}
input_exc_weights = network.connections['X', 'Y'].w
square_weights = get_square_weights(
input_exc_weights.view(784, n_neurons), n_sqrt, 28)
square_assignments = get_square_assignments(assignments, n_sqrt)
# inpt_axes, inpt_ims = plot_input(_input, reconstruction, label=labels[i], axes=inpt_axes, ims=inpt_ims)
spike_ims, spike_axes = plot_spikes(_spikes,
ims=spike_ims,
axes=spike_axes)
weights_im = plot_weights(square_weights, im=weights_im, wmax=0.25)
# assigns_im = plot_assignments(square_assignments, im=assigns_im)
# perf_ax = plot_performance(curves, ax=perf_ax)
plt.pause(1e-8)
network.reset_() # Reset state variables.
print(f'Progress: {n_examples} / {n_examples} ({t() - start:.4f} seconds)')
i += 1
if i % len(labels) == 0:
current_labels = labels[-update_interval:]
else:
current_labels = labels[i % len(images) - update_interval:i %
len(images)]
# Update and print accuracy evaluations.
curves, predictions = update_curves(curves,
current_labels,
n_classes,
spike_record=spike_record,
assignments=assignments,
proportions=proportions,
ngram_scores=ngram_scores,
n=2)
print_results(curves)
if train:
if any([x[-1] > best_accuracy for x in curves.values()]):
print('New best accuracy! Saving network parameters to disk.')
# Save network to disk.
if train:
path = os.path.join('..', '..', 'params', data, model)
if not os.path.isdir(path):
os.makedirs(path)
network.save(os.path.join(path, model_name + '.pt'))
path = os.path.join(
path, '_'.join(['auxiliary', model_name]) + '.pt')
torch.save((assignments, proportions, rates, ngram_scores),
open(path, 'wb'))
if train:
print('\nTraining complete.\n')
else:
print('\nTest complete.\n')
print('Average accuracies:\n')
for scheme in curves.keys():
print('\t%s: %.2f' % (scheme, float(np.mean(curves[scheme]))))
# Save accuracy curves to disk.
path = os.path.join('..', '..', 'curves', data, model)
if not os.path.isdir(path):
os.makedirs(path)
if train:
to_write = ['train'] + params
else:
to_write = ['test'] + params
to_write = [str(x) for x in to_write]
f = '_'.join(to_write) + '.pt'
torch.save((curves, update_interval, n_examples),
open(os.path.join(path, f), 'wb'))
# Save results to disk.
path = os.path.join('..', '..', 'results', data, model)
if not os.path.isdir(path):
os.makedirs(path)
results = [
np.mean(curves['all']),
np.mean(curves['proportion']),
np.mean(curves['ngram']),
np.max(curves['all']),
np.max(curves['proportion']),
np.max(curves['ngram'])
]
if train:
to_write = params + results
else:
to_write = test_params + results
to_write = [str(x) for x in to_write]
if train:
name = 'train.csv'
else:
name = 'test.csv'
if not os.path.isfile(os.path.join(path, name)):
with open(os.path.join(path, name), 'w') as f:
if train:
f.write(
'random_seed,n_neurons,n_train,inhib,time,lr,lr_decay,theta_plus,theta_decay,'
'progress_interval,update_interval,mean_all_activity,mean_proportion_weighting,'
'mean_ngram,max_all_activity,max_proportion_weighting,max_ngram\n'
)
else:
f.write(
'random_seed,n_neurons,n_train,n_test,inhib,time,lr,lr_decay,theta_plus,theta_decay,'
'progress_interval,update_interval,mean_all_activity,mean_proportion_weighting,'
'mean_ngram,max_all_activity,max_proportion_weighting,max_ngram\n'
)
with open(os.path.join(path, name), 'a') as f:
f.write(','.join(to_write) + '\n')
def main(args):
# Random seeding.
torch.manual_seed(args.seed)
# Device.
device = torch.device("cuda" if args.gpu else "cpu")
# No. workers.
if args.n_workers == -1:
args.n_workers = args.gpu * 4 * torch.cuda.device_count()
# Build network.
network = Network(batch_size=args.batch_size)
network.add_layer(Input(shape=(1, 28, 28), traces=True), name="I")
network.add_layer(LIFNodes(n=10,
traces=True,
rest=0,
reset=0,
thresh=1,
refrac=0),
name="O")
network.add_connection(
Connection(
source=network.layers["I"],
target=network.layers["O"],
nu=(0.0, 0.01),
update_rule=Hebbian,
wmin=0.0,
wmax=1.0,
norm=100.0,
reduction=torch.sum,
),
source="I",
target="O",
)
if args.plot:
for l in network.layers:
network.add_monitor(Monitor(network.layers[l],
state_vars=("s", ),
time=args.time),
name=l)
network.to(device)
# Load dataset.
dataset = MNIST(
image_encoder=PoissonEncoder(time=args.time, dt=1.0),
label_encoder=None,
root=os.path.join(ROOT_DIR, "data", "MNIST"),
download=True,
train=False,
transform=transforms.Compose(
[transforms.ToTensor(),
transforms.Lambda(lambda x: x * 250)]),
)
# Create a dataloader to iterate and batch data
dataloader = DataLoader(
dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.n_workers,
pin_memory=args.gpu,
)
spike_ims = None
spike_axes = None
weights_im = None
t0 = time()
for step, batch in enumerate(tqdm(dataloader)):
# Prep next input batch.
inputs = batch["encoded_image"]
inpts = {"I": inputs}
if args.gpu:
inpts = {k: v.cuda() for k, v in inpts.items()}
clamp = torch.nn.functional.one_hot(batch["label"],
num_classes=10).byte()
unclamp = ~clamp
clamp = {"O": clamp}
unclamp = {"O": unclamp}
# Run the network on the input.
network.run(
inpts=inpts,
time=args.time,
one_step=args.one_step,
clamp=clamp,
unclamp=unclamp,
)
if args.plot:
# Plot output spikes.
spikes = {
l: network.monitors[l].get("s")[:, 0]
for l in network.monitors
}
spike_ims, spike_axes = plot_spikes(spikes=spikes,
ims=spike_ims,
axes=spike_axes)
# Plot connection weights.
weights = network.connections["I", "O"].w
weights = get_square_weights(weights, n_sqrt=4, side=28)
weights_im = plot_weights(weights,
wmax=network.connections["I", "O"].wmax,
im=weights_im)
plt.pause(1e-2)
# Reset state variables.
network.reset_()
network.learning = False
for step, batch in enumerate(tqdm(dataloader)):
# Prep next input batch.
inputs = batch["encoded_image"]
inpts = {"I": inputs}
if args.gpu:
inpts = {k: v.cuda() for k, v in inpts.items()}
# Run the network on the input.
network.run(inpts=inpts, time=args.time, one_step=args.one_step)
if args.plot:
# Plot output spikes.
spikes = {
l: network.monitors[l].get("s")[:, 0]
for l in network.monitors
}
spike_ims, spike_axes = plot_spikes(spikes=spikes,
ims=spike_ims,
axes=spike_axes)
# Plot connection weights.
weights = network.connections["I", "O"].w
weights = get_square_weights(weights, n_sqrt=4, side=28)
weights_im = plot_weights(weights,
wmax=network.connections["I", "O"].wmax,
im=weights_im)
plt.pause(1e-2)
t1 = time() - t0
print(f"Time: {t1}")
def main(seed=0,
n_train=60000,
n_test=10000,
kernel_size=(16, ),
stride=(4, ),
n_filters=25,
padding=0,
inhib=100,
time=25,
lr=1e-3,
lr_decay=0.99,
dt=1,
intensity=1,
progress_interval=10,
update_interval=250,
plot=False,
train=True,
gpu=False):
assert n_train % update_interval == 0 and n_test % update_interval == 0, \
'No. examples must be divisible by update_interval'
params = [
seed, n_train, kernel_size, stride, n_filters, padding, inhib, time,
lr, lr_decay, dt, intensity, update_interval
]
model_name = '_'.join([str(x) for x in params])
if not train:
test_params = [
seed, n_train, n_test, kernel_size, stride, n_filters, padding,
inhib, time, lr, lr_decay, dt, intensity, update_interval
]
np.random.seed(seed)
if gpu:
torch.set_default_tensor_type('torch.cuda.FloatTensor')
torch.cuda.manual_seed_all(seed)
else:
torch.manual_seed(seed)
n_examples = n_train if train else n_test
input_shape = [20, 20]
if kernel_size == input_shape:
conv_size = [1, 1]
else:
conv_size = (int((input_shape[0] - kernel_size[0]) / stride[0]) + 1,
int((input_shape[1] - kernel_size[1]) / stride[1]) + 1)
n_classes = 10
n_neurons = n_filters * np.prod(conv_size)
total_kernel_size = int(np.prod(kernel_size))
total_conv_size = int(np.prod(conv_size))
# Build network.
if train:
network = Network()
input_layer = Input(n=400, shape=(1, 1, 20, 20), traces=True)
conv_layer = DiehlAndCookNodes(n=n_filters * total_conv_size,
shape=(1, n_filters, *conv_size),
thresh=-64.0,
traces=True,
theta_plus=0.05 * (kernel_size[0] / 20),
refrac=0)
conv_layer2 = LIFNodes(n=n_filters * total_conv_size,
shape=(1, n_filters, *conv_size),
refrac=0)
conv_conn = Conv2dConnection(input_layer,
conv_layer,
kernel_size=kernel_size,
stride=stride,
update_rule=WeightDependentPostPre,
norm=0.05 * total_kernel_size,
nu=[0, lr],
wmin=0,
wmax=0.25)
conv_conn2 = Conv2dConnection(input_layer,
conv_layer2,
w=conv_conn.w,
kernel_size=kernel_size,
stride=stride,
update_rule=None,
wmax=0.25)
w = -inhib * torch.ones(n_filters, conv_size[0], conv_size[1],
n_filters, conv_size[0], conv_size[1])
for f in range(n_filters):
for f2 in range(n_filters):
if f != f2:
w[f, :, :f2, :, :] = 0
w = w.view(n_filters * conv_size[0] * conv_size[1],
n_filters * conv_size[0] * conv_size[1])
recurrent_conn = Connection(conv_layer, conv_layer, w=w)
network.add_layer(input_layer, name='X')
network.add_layer(conv_layer, name='Y')
network.add_layer(conv_layer2, name='Y_')
network.add_connection(conv_conn, source='X', target='Y')
network.add_connection(conv_conn2, source='X', target='Y_')
network.add_connection(recurrent_conn, source='Y', target='Y')
# Voltage recording for excitatory and inhibitory layers.
voltage_monitor = Monitor(network.layers['Y'], ['v'], time=time)
network.add_monitor(voltage_monitor, name='output_voltage')
else:
network = load_network(os.path.join(params_path, model_name + '.pt'))
network.connections['X', 'Y'].update_rule = NoOp(
connection=network.connections['X', 'Y'],
nu=network.connections['X', 'Y'].nu)
network.layers['Y'].theta_decay = 0
network.layers['Y'].theta_plus = 0
# Load MNIST data.
dataset = MNIST(data_path, download=True)
if train:
images, labels = dataset.get_train()
else:
images, labels = dataset.get_test()
images *= intensity
images = images[:, 4:-4, 4:-4].contiguous()
# Record spikes during the simulation.
spike_record = torch.zeros(update_interval, time, n_neurons)
full_spike_record = torch.zeros(n_examples, n_neurons)
# Neuron assignments and spike proportions.
if train:
logreg_model = LogisticRegression(warm_start=True,
n_jobs=-1,
solver='lbfgs',
max_iter=1000,
multi_class='multinomial')
logreg_model.coef_ = np.zeros([n_classes, n_neurons])
logreg_model.intercept_ = np.zeros(n_classes)
logreg_model.classes_ = np.arange(n_classes)
else:
path = os.path.join(params_path,
'_'.join(['auxiliary', model_name]) + '.pt')
logreg_coef, logreg_intercept = torch.load(open(path, 'rb'))
logreg_model = LogisticRegression(warm_start=True,
n_jobs=-1,
solver='lbfgs',
max_iter=1000,
multi_class='multinomial')
logreg_model.coef_ = logreg_coef
logreg_model.intercept_ = logreg_intercept
logreg_model.classes_ = np.arange(n_classes)
# Sequence of accuracy estimates.
curves = {'logreg': []}
predictions = {scheme: torch.Tensor().long() for scheme in curves.keys()}
if train:
best_accuracy = 0
spikes = {}
for layer in set(network.layers):
spikes[layer] = Monitor(network.layers[layer],
state_vars=['s'],
time=time)
network.add_monitor(spikes[layer], name='%s_spikes' % layer)
# Train the network.
if train:
print('\nBegin training.\n')
else:
print('\nBegin test.\n')
inpt_ims = None
inpt_axes = None
spike_ims = None
spike_axes = None
weights_im = None
plot_update_interval = 100
start = t()
for i in range(n_examples):
if i % progress_interval == 0:
print('Progress: %d / %d (%.4f seconds)' %
(i, n_examples, t() - start))
start = t()
if i % update_interval == 0 and i > 0:
if train:
network.connections['X', 'Y'].update_rule.nu[1] *= lr_decay
if i % len(labels) == 0:
current_labels = labels[-update_interval:]
current_record = full_spike_record[-update_interval:]
else:
current_labels = labels[i % len(labels) - update_interval:i %
len(labels)]
current_record = full_spike_record[i % len(labels) -
update_interval:i %
len(labels)]
# Update and print accuracy evaluations.
curves, preds = update_curves(curves,
current_labels,
n_classes,
full_spike_record=current_record,
logreg=logreg_model)
print_results(curves)
for scheme in preds:
predictions[scheme] = torch.cat(
[predictions[scheme], preds[scheme]], -1)
# Save accuracy curves to disk.
to_write = ['train'] + params if train else ['test'] + params
f = '_'.join([str(x) for x in to_write]) + '.pt'
torch.save((curves, update_interval, n_examples),
open(os.path.join(curves_path, f), 'wb'))
if train:
if any([x[-1] > best_accuracy for x in curves.values()]):
print(
'New best accuracy! Saving network parameters to disk.'
)
# Save network to disk.
network.save(os.path.join(params_path, model_name + '.pt'))
path = os.path.join(
params_path,
'_'.join(['auxiliary', model_name]) + '.pt')
torch.save((logreg_model.coef_, logreg_model.intercept_),
open(path, 'wb'))
best_accuracy = max([x[-1] for x in curves.values()])
# Refit logistic regression model.
logreg_model = logreg_fit(full_spike_record[:i], labels[:i],
logreg_model)
print()
# Get next input sample.
image = images[i % len(images)]
sample = bernoulli(datum=image, time=time, dt=dt,
max_prob=1).unsqueeze(1).unsqueeze(1)
inpts = {'X': sample}
# Run the network on the input.
network.run(inpts=inpts, time=time)
network.connections['X', 'Y_'].w = network.connections['X', 'Y'].w
# Add to spikes recording.
spike_record[i % update_interval] = spikes['Y_'].get('s').view(
time, -1)
full_spike_record[i] = spikes['Y_'].get('s').view(time, -1).sum(0)
# Optionally plot various simulation information.
if plot and i % plot_update_interval == 0:
_input = inpts['X'].view(time, 400).sum(0).view(20, 20)
w = network.connections['X', 'Y'].w
_spikes = {
'X': spikes['X'].get('s').view(400, time),
'Y': spikes['Y'].get('s').view(n_filters * total_conv_size,
time),
'Y_': spikes['Y_'].get('s').view(n_filters * total_conv_size,
time)
}
inpt_axes, inpt_ims = plot_input(image.view(20, 20),
_input,
label=labels[i % len(labels)],
ims=inpt_ims,
axes=inpt_axes)
spike_ims, spike_axes = plot_spikes(spikes=_spikes,
ims=spike_ims,
axes=spike_axes)
weights_im = plot_conv2d_weights(
w, im=weights_im, wmax=network.connections['X', 'Y'].wmax)
plt.pause(1e-2)
network.reset_() # Reset state variables.
print(f'Progress: {n_examples} / {n_examples} ({t() - start:.4f} seconds)')
i += 1
if i % len(labels) == 0:
current_labels = labels[-update_interval:]
current_record = full_spike_record[-update_interval:]
else:
current_labels = labels[i % len(labels) - update_interval:i %
len(labels)]
current_record = full_spike_record[i % len(labels) -
update_interval:i % len(labels)]
# Update and print accuracy evaluations.
curves, preds = update_curves(curves,
current_labels,
n_classes,
full_spike_record=current_record,
logreg=logreg_model)
print_results(curves)
for scheme in preds:
predictions[scheme] = torch.cat([predictions[scheme], preds[scheme]],
-1)
if train:
if any([x[-1] > best_accuracy for x in curves.values()]):
print('New best accuracy! Saving network parameters to disk.')
# Save network to disk.
network.save(os.path.join(params_path, model_name + '.pt'))
path = os.path.join(params_path,
'_'.join(['auxiliary', model_name]) + '.pt')
torch.save((logreg_model.coef_, logreg_model.intercept_),
open(path, 'wb'))
if train:
print('\nTraining complete.\n')
else:
print('\nTest complete.\n')
print('Average accuracies:\n')
for scheme in curves.keys():
print('\t%s: %.2f' % (scheme, float(np.mean(curves[scheme]))))
# Save accuracy curves to disk.
to_write = ['train'] + params if train else ['test'] + params
to_write = [str(x) for x in to_write]
f = '_'.join(to_write) + '.pt'
torch.save((curves, update_interval, n_examples),
open(os.path.join(curves_path, f), 'wb'))
# Save results to disk.
results = [np.mean(curves['logreg']), np.std(curves['logreg'])]
to_write = params + results if train else test_params + results
to_write = [str(x) for x in to_write]
name = 'train.csv' if train else 'test.csv'
if not os.path.isfile(os.path.join(results_path, name)):
with open(os.path.join(results_path, name), 'w') as f:
if train:
columns = [
'seed', 'n_train', 'kernel_size', 'stride', 'n_filters',
'padding', 'inhib', 'time', 'lr', 'lr_decay', 'dt',
'intensity', 'update_interval', 'mean_logreg', 'std_logreg'
]
header = ','.join(columns) + '\n'
f.write(header)
else:
columns = [
'seed', 'n_train', 'n_test', 'kernel_size', 'stride',
'n_filters', 'padding', 'inhib', 'time', 'lr', 'lr_decay',
'dt', 'intensity', 'update_interval', 'mean_logreg',
'std_logreg'
]
header = ','.join(columns) + '\n'
f.write(header)
with open(os.path.join(results_path, name), 'a') as f:
f.write(','.join(to_write) + '\n')
if labels.numel() > n_examples:
labels = labels[:n_examples]
else:
while labels.numel() < n_examples:
if 2 * labels.numel() > n_examples:
labels = torch.cat(
[labels, labels[:n_examples - labels.numel()]])
else:
labels = torch.cat([labels, labels])
# Compute confusion matrices and save them to disk.
confusions = {}
for scheme in predictions:
confusions[scheme] = confusion_matrix(labels, predictions[scheme])
to_write = ['train'] + params if train else ['test'] + test_params
f = '_'.join([str(x) for x in to_write]) + '.pt'
torch.save(confusions, os.path.join(confusion_path, f))
inpt = Input(784, shape=(28, 28))
network.add_layer(inpt, name='I')
output = LIFNodes(625, thresh=-52 + torch.randn(625))
network.add_layer(output, name='O')
C1 = Connection(source=inpt, target=output, w=torch.randn(inpt.n, output.n))
C2 = Connection(source=output,
target=output,
w=0.5 * torch.randn(output.n, output.n))
network.add_connection(C1, source='I', target='O')
network.add_connection(C2, source='O', target='O')
spikes = {}
for l in network.layers:
spikes[l] = Monitor(network.layers[l], ['s'], time=250)
network.add_monitor(spikes[l], name='%s_spikes' % l)
voltages = {'O': Monitor(network.layers['O'], ['v'], time=250)}
network.add_monitor(voltages['O'], name='O_voltages')
# Get MNIST training images and labels.
images, labels = MNIST(path='../../data/MNIST', download=True).get_train()
images *= 0.25
# Create lazily iterating Poisson-distributed data loader.
loader = zip(poisson_loader(images, time=250), iter(labels))
inpt_axes = None
inpt_ims = None
spike_axes = None
spike_ims = None
def main(args):
if args.gpu:
torch.cuda.manual_seed_all(args.seed)
else:
torch.manual_seed(args.seed)
conv_size = int(
(28 - args.kernel_size + 2 * args.padding) / args.stride) + 1
# Build network.
network = Network()
input_layer = Input(n=784, shape=(1, 28, 28), traces=True)
conv_layer = DiehlAndCookNodes(
n=args.n_filters * conv_size * conv_size,
shape=(args.n_filters, conv_size, conv_size),
traces=True,
)
conv_conn = Conv2dConnection(
input_layer,
conv_layer,
kernel_size=args.kernel_size,
stride=args.stride,
update_rule=PostPre,
norm=0.4 * args.kernel_size**2,
nu=[0, args.lr],
reduction=max_without_indices,
wmax=1.0,
)
w = torch.zeros(args.n_filters, conv_size, conv_size, args.n_filters,
conv_size, conv_size)
for fltr1 in range(args.n_filters):
for fltr2 in range(args.n_filters):
if fltr1 != fltr2:
for i in range(conv_size):
for j in range(conv_size):
w[fltr1, i, j, fltr2, i, j] = -100.0
w = w.view(args.n_filters * conv_size * conv_size,
args.n_filters * conv_size * conv_size)
recurrent_conn = Connection(conv_layer, conv_layer, w=w)
network.add_layer(input_layer, name="X")
network.add_layer(conv_layer, name="Y")
network.add_connection(conv_conn, source="X", target="Y")
network.add_connection(recurrent_conn, source="Y", target="Y")
# Voltage recording for excitatory and inhibitory layers.
voltage_monitor = Monitor(network.layers["Y"], ["v"], time=args.time)
network.add_monitor(voltage_monitor, name="output_voltage")
if args.gpu:
network.to("cuda")
# Load MNIST data.
train_dataset = MNIST(
PoissonEncoder(time=args.time, dt=args.dt),
None,
os.path.join(ROOT_DIR, "data", "MNIST"),
download=True,
train=True,
transform=transforms.Compose([
transforms.ToTensor(),
transforms.Lambda(lambda x: x * args.intensity)
]),
)
spikes = {}
for layer in set(network.layers):
spikes[layer] = Monitor(network.layers[layer],
state_vars=["s"],
time=args.time)
network.add_monitor(spikes[layer], name="%s_spikes" % layer)
voltages = {}
for layer in set(network.layers) - {"X"}:
voltages[layer] = Monitor(network.layers[layer],
state_vars=["v"],
time=args.time)
network.add_monitor(voltages[layer], name="%s_voltages" % layer)
# Train the network.
print("Begin training.\n")
start = time()
weights_im = None
for epoch in range(args.n_epochs):
if epoch % args.progress_interval == 0:
print("Progress: %d / %d (%.4f seconds)" %
(epoch, args.n_epochs, time() - start))
start = time()
train_dataloader = DataLoader(
train_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=4,
pin_memory=args.gpu,
)
for step, batch in enumerate(tqdm(train_dataloader)):
# Get next input sample.
inpts = {"X": batch["encoded_image"]}
if args.gpu:
inpts = {k: v.cuda() for k, v in inpts.items()}
# Run the network on the input.
network.run(inpts=inpts, time=args.time, input_time_dim=0)
# Decay learning rate.
network.connections["X", "Y"].nu[1] *= 0.99
# Optionally plot various simulation information.
if args.plot:
weights = conv_conn.w
weights_im = plot_conv2d_weights(weights, im=weights_im)
plt.pause(1e-8)
network.reset_() # Reset state variables.
print("Progress: %d / %d (%.4f seconds)\n" %
(args.n_epochs, args.n_epochs, time() - start))
print("Training complete.\n")
def toLIF(network: Network):
new_network = Network(dt=1, learning=True)
input_layer = Input(n=network.X.n,
shape=network.X.shape,
traces=True,
tc_trace=network.X.tc_trace.item())
exc_layer = LIFNodes(
n=network.Ae.n,
traces=True,
rest=network.Ai.rest.item(),
reset=network.Ai.reset.item(),
thresh=network.Ai.thresh.item(),
refrac=network.Ai.refrac.item(),
tc_decay=network.Ai.tc_decay.item(),
)
inh_layer = LIFNodes(
n=network.Ai.n,
traces=False,
rest=network.Ai.rest.item(),
reset=network.Ai.reset.item(),
thresh=network.Ai.thresh.item(),
tc_decay=network.Ai.tc_decay.item(),
refrac=network.Ai.refrac.item(),
)
# Connections
w = network.X_to_Ae.w
input_exc_conn = Connection(
source=input_layer,
target=exc_layer,
w=w,
update_rule=PostPre,
nu=network.X_to_Ae.nu,
reduction=network.X_to_Ae.reduction,
wmin=network.X_to_Ae.wmin,
wmax=network.X_to_Ae.wmax,
norm=network.X_to_Ae.norm * 1,
)
w = network.Ae_to_Ai.w
exc_inh_conn = Connection(source=exc_layer,
target=inh_layer,
w=w,
wmin=network.Ae_to_Ai.wmin,
wmax=network.Ae_to_Ai.wmax)
w = network.Ai_to_Ae.w
inh_exc_conn = Connection(source=inh_layer,
target=exc_layer,
w=w,
wmin=network.Ai_to_Ae.wmin,
wmax=network.Ai_to_Ae.wmax)
# Add to network
new_network.add_layer(input_layer, name="X")
new_network.add_layer(exc_layer, name="Ae")
new_network.add_layer(inh_layer, name="Ai")
new_network.add_connection(input_exc_conn, source="X", target="Ae")
new_network.add_connection(exc_inh_conn, source="Ae", target="Ai")
new_network.add_connection(inh_exc_conn, source="Ai", target="Ae")
exc_voltage_monitor = Monitor(new_network.layers["Ae"], ["v"], time=500)
inh_voltage_monitor = Monitor(new_network.layers["Ai"], ["v"], time=500)
new_network.add_monitor(exc_voltage_monitor, name="exc_voltage")
new_network.add_monitor(inh_voltage_monitor, name="inh_voltage")
spikes = {}
for layer in set(network.layers):
spikes[layer] = Monitor(new_network.layers[layer],
state_vars=["s"],
time=time)
new_network.add_monitor(spikes[layer], name="%s_spikes" % layer)
return new_network
def main(seed=0, n_neurons=100, n_train=60000, n_test=10000, inhib=100, lr=0.01, lr_decay=1, time=350, dt=1,
theta_plus=0.05, theta_decay=1e-7, progress_interval=10, update_interval=250, plot=False,
train=True, gpu=False):
assert n_train % update_interval == 0 and n_test % update_interval == 0, \
'No. examples must be divisible by update_interval'
params = [
seed, n_neurons, n_train, inhib, lr_decay, time, dt,
theta_plus, theta_decay, progress_interval, update_interval
]
model_name = '_'.join([str(x) for x in params])
np.random.seed(seed)
if gpu:
torch.set_default_tensor_type('torch.cuda.FloatTensor')
torch.cuda.manual_seed_all(seed)
else:
torch.manual_seed(seed)
n_examples = n_train if train else n_test
n_classes = 10
# Build network.
if train:
network = Network(dt=dt)
input_layer = Input(n=784, traces=True, trace_tc=5e-2)
network.add_layer(input_layer, name='X')
output_layer = DiehlAndCookNodes(
n=n_classes, rest=0, reset=1, thresh=1, decay=1e-2,
theta_plus=theta_plus, theta_decay=theta_decay, traces=True, trace_tc=5e-2
)
network.add_layer(output_layer, name='Y')
w = torch.rand(784, n_classes)
input_connection = Connection(
source=input_layer, target=output_layer, w=w,
update_rule=MSTDPET, nu=lr, wmin=0, wmax=1,
norm=78.4, tc_e_trace=0.1
)
network.add_connection(input_connection, source='X', target='Y')
else:
network = load(os.path.join(params_path, model_name + '.pt'))
network.connections['X', 'Y'].update_rule = NoOp(
connection=network.connections['X', 'Y'], nu=network.connections['X', 'Y'].nu
)
network.layers['Y'].theta_decay = torch.IntTensor([0])
network.layers['Y'].theta_plus = torch.IntTensor([0])
# Load MNIST data.
environment = MNISTEnvironment(
dataset=MNIST(root=data_path, download=True), train=train, time=time
)
# Create pipeline.
pipeline = Pipeline(
network=network, environment=environment, encoding=repeat,
action_function=select_spiked, output='Y', reward_delay=None
)
spikes = {}
for layer in set(network.layers):
spikes[layer] = Monitor(network.layers[layer], state_vars=('s',), time=time)
network.add_monitor(spikes[layer], name='%s_spikes' % layer)
if train:
network.add_monitor(Monitor(
network.connections['X', 'Y'].update_rule, state_vars=('tc_e_trace',), time=time
), 'X_Y_e_trace')
# Train the network.
if train:
print('\nBegin training.\n')
else:
print('\nBegin test.\n')
spike_ims = None
spike_axes = None
weights_im = None
elig_axes = None
elig_ims = None
start = t()
for i in range(n_examples):
if i % progress_interval == 0:
print(f'Progress: {i} / {n_examples} ({t() - start:.4f} seconds)')
start = t()
if i > 0 and train:
network.connections['X', 'Y'].update_rule.nu[1] *= lr_decay
# Run the network on the input.
# print("Example",i,"Results:")
# for j in range(time):
# result = pipeline.env_step()
# pipeline.step(result,a_plus=1, a_minus=0)
# print(result)
for j in range(time):
pipeline.train()
if not train:
_spikes = {layer: spikes[layer].get('s') for layer in spikes}
if plot:
_spikes = {layer: spikes[layer].get('s') for layer in spikes}
w = network.connections['X', 'Y'].w
square_weights = get_square_weights(w.view(784, n_classes), 4, 28)
spike_ims, spike_axes = plot_spikes(_spikes, ims=spike_ims, axes=spike_axes)
weights_im = plot_weights(square_weights, im=weights_im)
elig_ims, elig_axes = plot_voltages(
{'Y': network.monitors['X_Y_e_trace'].get('e_trace').view(-1, time)[1500:2000]},
plot_type='line', ims=elig_ims, axes=elig_axes
)
plt.pause(1e-8)
pipeline.reset_state_variables() # Reset state variables.
network.connections['X', 'Y'].update_rule.tc_e_trace = torch.zeros(784, n_classes)
print(f'Progress: {n_examples} / {n_examples} ({t() - start:.4f} seconds)')
if train:
network.save(os.path.join(params_path, model_name + '.pt'))
print('\nTraining complete.\n')
else:
print('\nTest complete.\n')
# Determines number of workers to use
if n_workers == -1:
n_workers = gpu * 4 * torch.cuda.device_count()
print(n_workers)
print(torch.cuda.device_count())
n_sqrt = int(np.ceil(np.sqrt(n_neurons)))
start_intensity = intensity
network = Network(dt=1.0)
network.to(device)
network.to("cuda")
# state_vars: Iterable of strings indicating names of state variables to record.
for l in network.layers:
m = Monitor(network.layers[l], state_vars=['s'], time=time_arg)
network.add_monitor(m, name=l)
print(n_neurons, batch_size, nu_single, nu_pair)
# Build network.
network = DiehlAndCook2015(
n_inpt=784,
n_neurons=n_neurons,
exc=exc,
inh=inh,
dt=dt,
norm=78.4,
nu=(nu_single, nu_pair),
theta_plus=theta_plus,
inpt_shape=(1, 28, 28),
#emax=emax,
def forward(self, x):
return self.linear(x)
# Build a simple, two layer, "input-output" network.
network = Network(dt=1.0)
inpt = Input(784, shape=(28, 28))
network.add_layer(inpt, name='I')
output = LIFNodes(625, thresh=-52 + torch.randn(625))
network.add_layer(output, name='O')
network.add_connection(
Connection(inpt, output, w=torch.randn(inpt.n, output.n)), 'I', 'O')
network.add_connection(
Connection(output, output, w=0.5 * torch.randn(output.n, output.n)), 'O',
'O')
network.add_monitor(Monitor(output, ['s'], time=250), name='output_spikes')
# Get MNIST training images and labels and create data loader.
images, labels = MNIST(path='../../data/MNIST').get_train()
loader = zip(poisson_loader(images * 0.25, time=250), iter(labels))
# Run training data on reservoir and store (spikes per neuron, label) pairs.
training_pairs = []
for i, (datum, label) in enumerate(loader):
network.run(inpts={'I': datum}, time=250)
training_pairs.append(
[network.monitors['output_spikes'].get('s').sum(-1), label])
network.reset_()
if (i + 1) % 50 == 0: print('Train progress: (%d / 500)' % (i + 1))
if (i + 1) == 500:
def __init__(
self,
network: Network,
results_path: str,
dataset_name: str = "MNIST",
seed: int = 0,
batch_size: int = 1,
n_epochs: int = 1,
n_workers: int = -1,
update_interval: int = 250,
gif: bool = False,
gpu: bool = False,
plot: bool = False,
debug: bool = False,
) -> None:
"""
Constructor for class Spiking.
:param network: Network model to use.
:param results_path: Path to save training & testing results.
:param dataset_name: Name of dataset to use.
:param seed: Seed for pseudorandom number generator (PRNG).
:param batch_size: Mini-batch size.
:param n_epochs: Number of epochs for training.
:param n_workers: Number of workers to use.
:param update_interval: Interval to show network accuracy.
:param gif: Whether to create gif of weight maps.
:param gpu: Whether to use gpu.
:param plot: Whether to plot each timestep in real time
:param debug: Whether to save files for debugging purpose.
"""
self.network = network
self.results_path = results_path
self.batch_size = batch_size
self.n_epochs = n_epochs
self.n_workers = n_workers
self.update_interval = update_interval / batch_size
self.gif = gif
self.gpu = gpu
self.plot = plot
self.debug = debug
self.n_outpt = network.layers["Z"].n
self.profile = {
'method': network.method,
'dataset_name': dataset_name,
'n_train': None,
'n_test': None,
}
self.time = network.time
self.dt = network.dt
self.timestep = int(self.time / self.dt)
self.start_intensity_scale = 4
self.start_intensity = 256.0 / 8.0 * self.start_intensity_scale
self.train_dataset = None
self.validation_dataset = None
self.test_dataset = None
# Store results for checking purpose.
self.sl_train_spike = []
self.sl_test_spike = []
self.right_pred = []
self.wrong_pred = []
# Store network accuracy.
self.acc_history = {'train_acc': [], 'test_acc': []}
self.store_pred = {}
# Initialize plot class.
self.visualize = Plot()
# Save initial weights for plot.
self.exc_init_weight = network.connections[("X", "Y")].w.detach().clone()
self.sl_init_weight = network.connections[("Y", "Z")].w.detach().clone()
# Determines number of workers to use.
if n_workers == -1:
self.n_workers = gpu * 4 * torch.cuda.device_count()
# Sets max number of images to use for gif.
if gif:
self.n_gif_img = 35
# Sets up Gpu use.
if gpu:
torch.cuda.manual_seed_all(seed)
else:
torch.manual_seed(seed)
# Directs network to GPU.
if gpu:
network.to("cuda")
# Load train & test data.
encoder = PoissonEncoder(time=self.time, dt=self.dt)
self.train_dataset = load_data(dataset_name, encoder, True, self.start_intensity)
self.test_dataset = load_data(dataset_name, encoder, False, self.start_intensity)
# Set up monitors for spikes and voltages.
spikes = {}
for layer in set(network.layers):
l = network.layers[layer]
spikes[layer] = Monitor(l, state_vars=["s"], time=self.timestep)
network.add_monitor(spikes[layer], name="%s_spikes" % layer)
voltages = {}
for layer in set(network.layers) - {"X"}:
l = network.layers[layer]
voltages[layer] = Monitor(l, state_vars=["v"], time=self.timestep)
network.add_monitor(voltages[layer], name="%s_voltages" % layer)
#voltages['Y'] = Monitor(network.layers['Y'], state_vars=['theta'], time=self.timestep)
#network.add_monitor(voltages['Y'], name="Y_voltages")
self.spikes = spikes
self.voltages = voltages
def main(seed=0,
n_neurons=100,
n_train=60000,
n_test=10000,
c_low=1,
c_high=25,
p_low=0.5,
time=250,
dt=1,
theta_plus=0.05,
theta_decay=1e-7,
intensity=1,
progress_interval=10,
update_interval=250,
plot=False,
train=True,
gpu=False):
assert n_train % update_interval == 0 and n_test % update_interval == 0,\
'No. examples must be divisible by update_interval'
params = [
seed, n_neurons, n_train, c_low, c_high, p_low, time, dt, theta_plus,
theta_decay, intensity, progress_interval, update_interval
]
model_name = '_'.join([str(x) for x in params])
if not train:
test_params = [
seed, n_neurons, n_train, n_test, c_low, c_high, p_low, time, dt,
theta_plus, theta_decay, intensity, progress_interval,
update_interval
]
np.random.seed(seed)
if gpu:
torch.set_default_tensor_type('torch.cuda.FloatTensor')
torch.cuda.manual_seed_all(seed)
else:
torch.manual_seed(seed)
n_examples = n_train if train else n_test
n_sqrt = int(np.ceil(np.sqrt(n_neurons)))
n_classes = 10
# Build network.
if train:
network = Network(dt=dt)
input_layer = Input(n=784, traces=True)
exc_layer = DiehlAndCookNodes(n=n_neurons, traces=True)
w = torch.rand(input_layer.n, exc_layer.n)
input_exc_conn = Connection(input_layer,
exc_layer,
w=w,
update_rule=PostPre,
norm=78.4,
nu=(1e-4, 1e-2),
wmax=1.0)
w = torch.zeros(exc_layer.n, exc_layer.n)
for k1 in range(n_neurons):
for k2 in range(n_neurons):
if k1 != k2:
x1, y1 = k1 // np.sqrt(n_neurons), k1 % np.sqrt(n_neurons)
x2, y2 = k2 // np.sqrt(n_neurons), k2 % np.sqrt(n_neurons)
w[k1, k2] = max(
-c_high,
-c_low * np.sqrt(euclidean([x1, y1], [x2, y2])))
recurrent_conn = Connection(exc_layer, exc_layer, w=w)
network.add_layer(input_layer, name='X')
network.add_layer(exc_layer, name='Y')
network.add_connection(input_exc_conn, source='X', target='Y')
network.add_connection(recurrent_conn, source='Y', target='Y')
else:
network = load_network(os.path.join(params_path, model_name + '.pt'))
network.connections['X', 'Y'].update_rule = NoOp(
connection=network.connections['X', 'Y'],
nu=network.connections['X', 'Y'].nu)
network.layers['Y'].theta_decay = 0
network.layers['Y'].theta_plus = 0
# Load MNIST data.
dataset = MNIST(data_path, download=True)
if train:
images, labels = dataset.get_train()
else:
images, labels = dataset.get_test()
images = images.view(-1, 784)
images *= intensity
# Record spikes during the simulation.
spike_record = torch.zeros(update_interval, int(time / dt), n_neurons)
# Neuron assignments and spike proportions.
if train:
assignments = -torch.ones_like(torch.Tensor(n_neurons))
proportions = torch.zeros_like(torch.Tensor(n_neurons, 10))
rates = torch.zeros_like(torch.Tensor(n_neurons, 10))
ngram_scores = {}
else:
path = os.path.join(params_path,
'_'.join(['auxiliary', model_name]) + '.pt')
assignments, proportions, rates, ngram_scores = torch.load(
open(path, 'rb'))
# Sequence of accuracy estimates.
curves = {'all': [], 'proportion': [], 'ngram': []}
predictions = {scheme: torch.Tensor().long() for scheme in curves.keys()}
if train:
best_accuracy = 0
spikes = {}
for layer in set(network.layers) - {'X'}:
spikes[layer] = Monitor(network.layers[layer],
state_vars=['s'],
time=int(time / dt))
network.add_monitor(spikes[layer], name='%s_spikes' % layer)
# Train the network.
if train:
print('\nBegin training.\n')
else:
print('\nBegin test.\n')
inpt_axes = None
inpt_ims = None
spike_ims = None
spike_axes = None
weights_im = None
assigns_im = None
perf_ax = None
# Calculate linear increase every update interval.
if train:
n_increase = int(p_low * n_examples) / update_interval
increase = (c_high - c_low) / n_increase
increases = 0
inhib = c_low
start = t()
for i in range(n_examples):
if train and i % update_interval == 0 and i > 0 and increases < n_increase:
inhib = inhib + increase
print(f'\nIncreasing inhibition to {inhib}.\n')
w = torch.zeros(n_neurons, n_neurons)
for k1 in range(n_neurons):
for k2 in range(n_neurons):
if k1 != k2:
x1, y1 = k1 // np.sqrt(n_neurons), k1 % np.sqrt(
n_neurons)
x2, y2 = k2 // np.sqrt(n_neurons), k2 % np.sqrt(
n_neurons)
w[k1, k2] = max(
-c_high,
-inhib * np.sqrt(euclidean([x1, y1], [x2, y2])))
network.connections['Y', 'Y'].w = w
increases += 1
if i % progress_interval == 0:
print(f'Progress: {i} / {n_examples} ({t() - start:.4f} seconds)')
start = t()
if i % update_interval == 0 and i > 0:
if i % len(labels) == 0:
current_labels = labels[-update_interval:]
else:
current_labels = labels[i % len(images) - update_interval:i %
len(images)]
# Update and print accuracy evaluations.
curves, preds = update_curves(curves,
current_labels,
n_classes,
spike_record=spike_record,
assignments=assignments,
proportions=proportions,
ngram_scores=ngram_scores,
n=2)
print_results(curves)
for scheme in preds:
predictions[scheme] = torch.cat(
[predictions[scheme], preds[scheme]], -1)
# Save accuracy curves to disk.
to_write = ['train'] + params if train else ['test'] + params
f = '_'.join([str(x) for x in to_write]) + '.pt'
torch.save((curves, update_interval, n_examples),
open(os.path.join(curves_path, f), 'wb'))
if train:
if any([x[-1] > best_accuracy for x in curves.values()]):
print(
'New best accuracy! Saving network parameters to disk.'
)
# Save network to disk.
network.save(os.path.join(params_path, model_name + '.pt'))
path = os.path.join(
params_path,
'_'.join(['auxiliary', model_name]) + '.pt')
torch.save((assignments, proportions, rates, ngram_scores),
open(path, 'wb'))
best_accuracy = max([x[-1] for x in curves.values()])
# Assign labels to excitatory layer neurons.
assignments, proportions, rates = assign_labels(
spike_record, labels[i - update_interval:i], 10, rates)
# Compute ngram scores.
ngram_scores = update_ngram_scores(
spike_record, labels[i - update_interval:i], 10, 2,
ngram_scores)
print()
# Get next input sample.
image = images[i]
sample = poisson(datum=image, time=int(time / dt))
inpts = {'X': sample}
# Run the network on the input.
network.run(inpts=inpts, time=time)
retries = 0
while spikes['Y'].get('s').sum() < 5 and retries < 3:
retries += 1
image *= 2
sample = poisson(datum=image, time=int(time / dt))
inpts = {'X': sample}
network.run(inpts=inpts, time=time)
# Add to spikes recording.
spike_record[i % update_interval] = spikes['Y'].get('s').t()
# Optionally plot various simulation information.
if plot:
inpt = inpts['X'].view(time, 784).sum(0).view(28, 28)
_spikes = {layer: spikes[layer].get('s') for layer in spikes}
input_exc_weights = network.connections['X', 'Y'].w
square_weights = get_square_weights(
input_exc_weights.view(784, n_neurons), n_sqrt, 28)
square_assignments = get_square_assignments(assignments, n_sqrt)
# inpt_axes, inpt_ims = plot_input(images[i].view(28, 28), inpt, label=labels[i], axes=inpt_axes, ims=inpt_ims)
spike_ims, spike_axes = plot_spikes(_spikes,
ims=spike_ims,
axes=spike_axes)
weights_im = plot_weights(square_weights, im=weights_im)
# assigns_im = plot_assignments(square_assignments, im=assigns_im)
# perf_ax = plot_performance(curves, ax=perf_ax)
plt.pause(1e-8)
network.reset_() # Reset state variables.
print(f'Progress: {n_examples} / {n_examples} ({t() - start:.4f} seconds)')
i += 1
if i % len(labels) == 0:
current_labels = labels[-update_interval:]
else:
current_labels = labels[i % len(images) - update_interval:i %
len(images)]
# Update and print accuracy evaluations.
curves, preds = update_curves(curves,
current_labels,
n_classes,
spike_record=spike_record,
assignments=assignments,
proportions=proportions,
ngram_scores=ngram_scores,
n=2)
print_results(curves)
for scheme in preds:
predictions[scheme] = torch.cat([predictions[scheme], preds[scheme]],
-1)
if train:
if any([x[-1] > best_accuracy for x in curves.values()]):
print('New best accuracy! Saving network parameters to disk.')
# Save network to disk.
network.save(os.path.join(params_path, model_name + '.pt'))
path = os.path.join(params_path,
'_'.join(['auxiliary', model_name]) + '.pt')
torch.save((assignments, proportions, rates, ngram_scores),
open(path, 'wb'))
if train:
print('\nTraining complete.\n')
else:
print('\nTest complete.\n')
print('Average accuracies:\n')
for scheme in curves.keys():
print(f'\t%s: %.2f' % (scheme, float(np.mean(curves[scheme]))))
# Save accuracy curves to disk.
to_write = ['train'] + params if train else ['test'] + params
to_write = [str(x) for x in to_write]
f = '_'.join(to_write) + '.pt'
torch.save((curves, update_interval, n_examples),
open(os.path.join(curves_path, f), 'wb'))
results = [
np.mean(curves['all']),
np.mean(curves['proportion']),
np.mean(curves['ngram']),
np.max(curves['all']),
np.max(curves['proportion']),
np.max(curves['ngram'])
]
to_write = params + results if train else test_params + results
to_write = [str(x) for x in to_write]
name = 'train.csv' if train else 'test.csv'
if not os.path.isfile(os.path.join(results_path, name)):
with open(os.path.join(results_path, name), 'w') as f:
if train:
f.write(
'random_seed,n_neurons,n_train,excite,c_low,c_high,p_low,time,timestep,theta_plus,theta_decay,'
'intensity,progress_interval,update_interval,mean_all_activity,mean_proportion_weighting,'
'mean_ngram,max_all_activity,max_proportion_weighting,max_ngram\n'
)
else:
f.write(
'random_seed,n_neurons,n_train,n_test,excite,c_low,c_high,p_low,time,timestep,theta_plus,theta_decay,'
'intensity,progress_interval,update_interval,mean_all_activity,mean_proportion_weighting,'
'mean_ngram,max_all_activity,max_proportion_weighting,max_ngram\n'
)
with open(os.path.join(results_path, name), 'a') as f:
f.write(','.join(to_write) + '\n')
if labels.numel() > n_examples:
labels = labels[:n_examples]
else:
while labels.numel() < n_examples:
if 2 * labels.numel() > n_examples:
labels = torch.cat(
[labels, labels[:n_examples - labels.numel()]])
else:
labels = torch.cat([labels, labels])
# Compute confusion matrices and save them to disk.
confusions = {}
for scheme in predictions:
confusions[scheme] = confusion_matrix(labels, predictions[scheme])
to_write = ['train'] + params if train else ['test'] + test_params
f = '_'.join([str(x) for x in to_write]) + '.pt'
torch.save(confusions, os.path.join(confusion_path, f))
print()
source="PK_Anti",
target="DCN_Anti")
GR_monitor = Monitor(obj=GR_Joint_layer, state_vars=("s"), time=time)
PK_monitor = Monitor(obj=PK, state_vars=("s", "v"), time=time)
PK_Anti_monitor = Monitor(obj=PK_Anti, state_vars=("s", "v"), time=time)
IO_monitor = Monitor(obj=IO, state_vars=("s"), time=time)
DCN_monitor = Monitor(
obj=DCN,
state_vars=("s", "v"),
time=time,
)
DCN_Anti_monitor = Monitor(obj=DCN_Anti, state_vars=("s", "v"), time=time)
network.add_monitor(monitor=GR_monitor, name="GR")
network.add_monitor(monitor=PK_monitor, name="PK")
network.add_monitor(monitor=PK_Anti_monitor, name="PK_Anti")
network.add_monitor(monitor=IO_monitor, name="IO")
network.add_monitor(monitor=DCN_monitor, name="DCN")
network.add_monitor(monitor=DCN_Anti_monitor, name="DCN_Anti")
data_Joint = torch.bernoulli(0.01 * torch.rand(time, GR_Joint_layer.n)).byte()
# 随机生成 error(无时间维度）
error = 0.1 * torch.rand(IO.n)
Curr_list = Error2IO_Current(error)
print(Curr_list)
IO_Input = poisson(Curr_list[0], time=time)
print(IO_Input)
IO_Anti_Input = poisson(Curr_list[1], time=time)
inputs = {
# learning rule
KC_EN.update_rule = STDP(connection=KC_EN,
nu=(-A, -A),
tc_eligibility_trace=40.0,
tc_plus=15,
tc_minus=15,
tc_reward=20.0,
min_weight=min_weight)
# monitors
input_monitor = Monitor(obj=input_layer, state_vars=("s"))
PN_monitor = Monitor(obj=PN, state_vars=("s", "v"))
KC_monitor = Monitor(obj=KC, state_vars=("s", "v"))
EN_monitor = Monitor(obj=EN, state_vars=("s", "v"))
landmark_guidance.add_monitor(monitor=input_monitor, name="Input monitor")
landmark_guidance.add_monitor(monitor=PN_monitor, name="PN monitor")
landmark_guidance.add_monitor(monitor=KC_monitor, name="KC monitor")
landmark_guidance.add_monitor(monitor=EN_monitor, name="EN monitor")
print("Run")
for phase in [1, 2, 3]:
print("Running phase", phase)
landmark_guidance.reset_state_variables()
if phase == 2:
landmark_guidance.learning = True
else:
landmark_guidance.learning = False
labels = torch.empty(1, dtype=torch.int)
# create a spike monitor for each layer in the network
# this allows us to read the spikes in order to assign labels to neurons and determine the predicted class
layer_monitors = {}
for layer in set(network.layers):
# initialize spike monitor at the layer
# do not record the voltage if at the input layer
state_vars = ["s", "v"] if (layer != input_layer_name) else ["s"]
layer_monitors[layer] = Monitor(network.layers[layer],
state_vars=state_vars,
time=int(time / dt))
# connect the monitor to the network
network.add_monitor(layer_monitors[layer], name="%s_spikes" % layer)
weight_history = None
num_correct = 0.0
### DEBUG ###
### can be used to force the network to learn the inputs in a specific way
supervised = True
### used to determine if status messages are printed out at each sample
log_messages = False
### used to show weight changes
graph_weights = False
###############
# show current weights
#print("Current Weights:")
# Voltage recordings for excitatory and readout layers.
voltages = {}
for layer in set(layers.keys()) - {'X'}:
voltages[layer] = Monitor(layers[layer], ['v'], time=plot_interval)
# Add all layers and connections to the network.
for layer in layers:
network.add_layer(layers[layer], name=layer)
network.add_connection(input_exc_conn, source='X', target='E')
network.add_connection(exc_readout_conn, source='E', target='R')
# Add all monitors to the network.
for layer in layers:
network.add_monitor(spikes[layer], name='%s_spikes' % layer)
if layer in voltages:
network.add_monitor(voltages[layer], name='%s_voltages' % layer)
# Load SpaceInvaders environment.
environment = GymEnvironment('Asteroids-v0')
environment.reset()
pipeline = Pipeline(network, environment, encoding=bernoulli, time=1, history=5, delta=10, plot_interval=plot_interval,
print_interval=print_interval, render_interval=render_interval, action_function=select_multinomial,
output='R')
total = 0
rewards = []
avg_rewards = []
def main(seed=0,
n_neurons=100,
n_train=60000,
n_test=10000,
inhib=250,
lr=1e-2,
lr_decay=1,
time=100,
dt=1,
theta_plus=0.05,
theta_decay=1e-7,
intensity=1,
progress_interval=10,
update_interval=100,
plot=False,
train=True,
gpu=False,
no_inhib=False,
no_theta=False):
assert n_train % update_interval == 0 and n_test % update_interval == 0, \
'No. examples must be divisible by update_interval'
params = [
seed, n_neurons, n_train, inhib, lr, lr_decay, time, dt, theta_plus,
theta_decay, intensity, progress_interval, update_interval
]
test_params = [
seed, n_neurons, n_train, n_test, inhib, lr, lr_decay, time, dt,
theta_plus, theta_decay, intensity, progress_interval, update_interval
]
model_name = '_'.join([str(x) for x in params])
np.random.seed(seed)
if gpu:
torch.set_default_tensor_type('torch.cuda.FloatTensor')
torch.cuda.manual_seed_all(seed)
else:
torch.manual_seed(seed)
n_examples = n_train if train else n_test
n_sqrt = int(np.ceil(np.sqrt(n_neurons)))
n_classes = 10
# Build network.
if train:
network = Network()
input_layer = Input(n=784, traces=True, trace_tc=5e-2)
network.add_layer(input_layer, name='X')
output_layer = DiehlAndCookNodes(n=n_neurons,
traces=True,
rest=0,
reset=0,
thresh=5,
refrac=0,
decay=1e-2,
trace_tc=5e-2,
theta_plus=theta_plus,
theta_decay=theta_decay)
network.add_layer(output_layer, name='Y')
w = 0.3 * torch.rand(784, n_neurons)
input_connection = Connection(source=network.layers['X'],
target=network.layers['Y'],
w=w,
update_rule=WeightDependentPostPre,
nu=[0, lr],
wmin=0,
wmax=1,
norm=78.4)
network.add_connection(input_connection, source='X', target='Y')
w = -inhib * (torch.ones(n_neurons, n_neurons) -
torch.diag(torch.ones(n_neurons)))
recurrent_connection = Connection(source=network.layers['Y'],
target=network.layers['Y'],
w=w,
wmin=-inhib,
wmax=0,
update_rule=WeightDependentPostPre,
nu=[0, -100 * lr],
norm=inhib / 2 * n_neurons)
network.add_connection(recurrent_connection, source='Y', target='Y')
mask = network.connections['Y', 'Y'].w == 0
masks = {('Y', 'Y'): mask}
else:
network = load_network(os.path.join(params_path, model_name + '.pt'))
network.connections['X', 'Y'].update_rule = NoOp(
connection=network.connections['X', 'Y'],
nu=network.connections['X', 'Y'].nu)
network.connections['Y', 'Y'].update_rule = NoOp(
connection=network.connections['X', 'Y'],
nu=network.connections['X', 'Y'].nu)
network.layers['Y'].theta_decay = 0
network.layers['Y'].theta_plus = 0
if no_inhib:
del network.connections['Y', 'Y']
if no_theta:
network.layers['Y'].theta = 0
# Load MNIST data.
dataset = MNIST(path=data_path, download=True)
if train:
images, labels = dataset.get_train()
else:
images, labels = dataset.get_test()
images = images.view(-1, 784)
images *= intensity
labels = labels.long()
monitors = {}
for layer in set(network.layers):
if 'v' in network.layers[layer].__dict__:
monitors[layer] = Monitor(network.layers[layer],
state_vars=['s', 'v'],
time=time)
else:
monitors[layer] = Monitor(network.layers[layer],
state_vars=['s'],
time=time)
network.add_monitor(monitors[layer], name=layer)
# Train the network.
if train:
print('\nBegin training.\n')
else:
print('\nBegin test.\n')
inpt_axes = None
inpt_ims = None
spike_ims = None
spike_axes = None
voltage_ims = None
voltage_axes = None
weights_im = None
weights2_im = None
unclamps = {}
per_class = int(n_neurons / n_classes)
for label in range(n_classes):
unclamp = torch.ones(n_neurons).byte()
unclamp[label * per_class:(label + 1) * per_class] = 0
unclamps[label] = unclamp
predictions = torch.zeros(n_examples)
corrects = torch.zeros(n_examples)
start = t()
for i in range(n_examples):
if i % progress_interval == 0:
print(f'Progress: {i} / {n_examples} ({t() - start:.4f} seconds)')
start = t()
if i % update_interval == 0 and i > 0 and train:
network.save(os.path.join(params_path, model_name + '.pt'))
network.connections['X', 'Y'].update_rule.nu[1] *= lr_decay
# Get next input sample.
image = images[i % len(images)]
label = labels[i % len(images)].item()
sample = poisson(datum=image, time=time, dt=dt)
inpts = {'X': sample}
# Run the network on the input.
if train:
network.run(inpts=inpts,
time=time,
unclamp={'Y': unclamps[label]},
masks=masks)
else:
network.run(inpts=inpts, time=time)
if not train:
retries = 0
while monitors['Y'].get('s').sum() == 0 and retries < 3:
retries += 1
image *= 1.5
sample = poisson(datum=image, time=time, dt=dt)
inpts = {'X': sample}
if train:
network.run(inpts=inpts,
time=time,
unclamp={'Y': unclamps[label]},
masks=masks)
else:
network.run(inpts=inpts, time=time)
output = monitors['Y'].get('s')
summed_neurons = output.sum(dim=1).view(n_classes, per_class)
summed_classes = summed_neurons.sum(dim=1)
prediction = torch.argmax(summed_classes).item()
correct = prediction == label
predictions[i] = prediction
corrects[i] = int(correct)
# Optionally plot various simulation information.
if plot:
# _input = image.view(28, 28)
# reconstruction = inpts['X'].view(time, 784).sum(0).view(28, 28)
# v = {'Y': monitors['Y'].get('v')}
s = {layer: monitors[layer].get('s') for layer in monitors}
input_exc_weights = network.connections['X', 'Y'].w
square_weights = get_square_weights(
input_exc_weights.view(784, n_neurons), n_sqrt, 28)
recurrent_weights = network.connections['Y', 'Y'].w
# inpt_axes, inpt_ims = plot_input(_input, reconstruction, label=labels[i], axes=inpt_axes, ims=inpt_ims)
# voltage_ims, voltage_axes = plot_voltages(v, ims=voltage_ims, axes=voltage_axes)
spike_ims, spike_axes = plot_spikes(s,
ims=spike_ims,
axes=spike_axes)
weights_im = plot_weights(square_weights, im=weights_im)
weights2_im = plot_weights(recurrent_weights,
im=weights2_im,
wmin=-inhib,
wmax=0)
plt.pause(1e-8)
network.reset_() # Reset state variables.
print(f'Progress: {n_examples} / {n_examples} ({t() - start:.4f} seconds)')
if train:
network.save(os.path.join(params_path, model_name + '.pt'))
if train:
print('\nTraining complete.\n')
else:
print('\nTest complete.\n')
accuracy = torch.mean(corrects).item() * 100
print(f'\nAccuracy: {accuracy}\n')
to_write = params + [accuracy] if train else test_params + [accuracy]
to_write = [str(x) for x in to_write]
name = 'train.csv' if train else 'test.csv'
if not os.path.isfile(os.path.join(results_path, name)):
with open(os.path.join(results_path, name), 'w') as f:
if train:
f.write(
'random_seed,n_neurons,n_train,inhib,lr,lr_decay,time,timestep,theta_plus,'
'theta_decay,intensity,progress_interval,update_interval,accuracy\n'
)
else:
f.write(
'random_seed,n_neurons,n_train,n_test,inhib,lr,lr_decay,time,timestep,'
'theta_plus,theta_decay,intensity,progress_interval,update_interval,accuracy\n'
)
with open(os.path.join(results_path, name), 'a') as f:
f.write(','.join(to_write) + '\n')
if labels.numel() > n_examples:
labels = labels[:n_examples]
else:
while labels.numel() < n_examples:
if 2 * labels.numel() > n_examples:
labels = torch.cat(
[labels, labels[:n_examples - labels.numel()]])
else:
labels = torch.cat([labels, labels])
# Compute confusion matrices and save them to disk.
confusion = confusion_matrix(labels, predictions)
to_write = ['train'] + params if train else ['test'] + test_params
f = '_'.join([str(x) for x in to_write]) + '.pt'
torch.save(confusion, os.path.join(confusion_path, f))
inpt = Input(784, shape=(28, 28))
network.add_layer(inpt, name="I")
output = LIFNodes(625, thresh=-52 + torch.randn(625))
network.add_layer(output, name="O")
C1 = Connection(source=inpt, target=output, w=torch.randn(inpt.n, output.n))
C2 = Connection(source=output,
target=output,
w=0.5 * torch.randn(output.n, output.n))
network.add_connection(C1, source="I", target="O")
network.add_connection(C2, source="O", target="O")
spikes = {}
for l in network.layers:
spikes[l] = Monitor(network.layers[l], ["s"], time=250)
network.add_monitor(spikes[l], name="%s_spikes" % l)
voltages = {"O": Monitor(network.layers["O"], ["v"], time=250)}
network.add_monitor(voltages["O"], name="O_voltages")
# Get MNIST training images and labels.
images, labels = MNIST(path="../../data/MNIST", download=True).get_train()
images *= 0.25
# Create lazily iterating Poisson-distributed data loader.
loader = zip(poisson_loader(images, time=250), iter(labels))
inpt_axes = None
inpt_ims = None
spike_axes = None
spike_ims = None
network = Network(dt=1.0)
inpt = Input(n=784, traces=True)
exc = AdaptiveLIFNodes(n=n, traces=True)
ew = 0.3 * torch.rand(784, n)
econn = Connection(inpt, exc, w=ew, update_rule=MSTDPET, nu=0.1, wmin=0, wmax=1, norm=78.4)
network.add_layer(inpt, 'X')
network.add_layer(exc, 'Y')
network.add_connection(econn, source='X', target='Y')
spike_monitors = {layer : Monitor(network.layers[layer], ['s']) for layer in network.layers}
for layer in spike_monitors:
network.add_monitor(spike_monitors[layer], '%s' % layer)
# Load MNIST data.
images, labels = MNIST(path=os.path.join('..', '..', 'data', 'MNIST'),
download=True).get_train()
images *= intensity
images /= 4
# Lazily encode data as Poisson spike trains.
ims = []
for j in range(0, iters, change_interval):
ims.extend([images[j % 60000]] * change_interval)
lbls = []
for j in range(0, iters, change_interval):
lbls.extend([int(labels[j % 60000])] * change_interval)
network.add_connection(FF2a, source="TNN_1a", target="rTNN_1")
network.add_connection(FF2b, source="TNN_1b", target="rTNN_1")
# (Recurrences)
network.add_connection(rTNN_to_buf1, source="rTNN_1", target="BUF_1")
# network.add_connection(buf1_to_buf2, source="BUF_1", target="BUF_2")
network.add_connection(buf1_to_rTNN, source="BUF_1", target="rTNN_1")
# network.add_connection(buf2_to_rTNN, source="BUF_2", target="rTNN_1")
# End of network creation
# Monitors:
spikes = {}
for l in network.layers:
spikes[l] = Monitor(network.layers[l], ["s"], time=num_timesteps)
network.add_monitor(spikes[l], name="%s_spikes" % l)
# Data and initial encoding:
dataset = MNIST(
RampNoLeakTNNEncoder(time=num_timesteps, dt=1),
None,
root=os.path.join("..", "..", "data", "MNIST"),
download=True,
transform=transforms.Compose(
[transforms.ToTensor(), transforms.Lambda(lambda x: x * intensity)]
),
)
# Create a dataloader to iterate and batch data
)
mstdp_1 = Connection(
source=inpt_mstdp, target=hiddn_mstdp, update_rule=MSTDP, wmin=w_min_1, wmax=w_max_1, nu=gamma_mstdp
)
mstdp_2 = Connection(
source=hiddn_mstdp, target=outpt_mstdp, update_rule=MSTDP, wmin=w_min_2, wmax=w_max_2, nu=gamma_mstdp
)
network_mstdp = Network(dt=dt)
network_mstdp.add_layer(name='Input', layer=inpt_mstdp)
network_mstdp.add_layer(name='Hidden', layer=hiddn_mstdp)
network_mstdp.add_layer(name='Output', layer=outpt_mstdp)
network_mstdp.add_connection(source='Input', target='Hidden', connection=mstdp_1)
network_mstdp.add_connection(source='Hidden', target='Output', connection=mstdp_2)
network_mstdp.add_monitor(name='In', monitor=Monitor(obj=inpt_mstdp, state_vars=['s'], time=100))
network_mstdp.add_monitor(name='Hid', monitor=Monitor(obj=hiddn_mstdp, state_vars=['s', 'v'], time=100))
# MSTDPET
torch.manual_seed(seed)
np.random.seed(seed)
inpt_mstdpet = Input(n_in)
hiddn_mstdpet = LIFNodes(
n_hidden, thresh=thresh_lif, rest=rest_lif, reset=reset_lif, tc_decay=tau_lif, refrac=refrac_lif
)
outpt_mstdpet = LIFNodes(
n_out, thresh=thresh_lif, rest=rest_lif, reset=reset_lif, tc_decay=tau_lif, refrac=refrac_lif
)
mstdpet_1 = Connection(
for i in range(conv_size):
for j in range(conv_size):
w[fltr1, i, j, fltr2, i, j] = -100.0
w = w.view(n_filters * conv_size * conv_size,
n_filters * conv_size * conv_size)
recurrent_conn = Connection(conv_layer, conv_layer, w=w)
network.add_layer(input_layer, name="X")
network.add_layer(conv_layer, name="Y")
network.add_connection(conv_conn, source="X", target="Y")
network.add_connection(recurrent_conn, source="Y", target="Y")
# Voltage recording for excitatory and inhibitory layers.
voltage_monitor = Monitor(network.layers["Y"], ["v"], time=time)
network.add_monitor(voltage_monitor, name="output_voltage")
if gpu:
network.to("cuda")
# Load MNIST data.
train_dataset = MNIST(
PoissonEncoder(time=time, dt=dt),
None,
"../../data/MNIST",
download=True,
train=True,
transform=transforms.Compose(
[transforms.ToTensor(),
transforms.Lambda(lambda x: x * intensity)]),
)
network = Network()
X = Input(n=100)
Y = LIFNodes(n=100)
C = MeanFieldConnection(source=X, target=Y, norm=100.0)
M_X = Monitor(X, state_vars=['s'])
M_Y = Monitor(Y, state_vars=['s', 'v'])
M_C = Monitor(C, state_vars=['w'])
network.add_layer(X, name='X')
network.add_layer(Y, name='Y')
network.add_connection(C, source='X', target='Y')
network.add_monitor(M_X, 'M_X')
network.add_monitor(M_Y, 'M_Y')
network.add_monitor(M_C, 'M_C')
spikes = torch.bernoulli(torch.rand(1000, 100))
inpts = {'X': spikes}
network.run(inpts=inpts, time=1000)
spikes = {'X': M_X.get('s'), 'Y': M_Y.get('s')}
weights = M_C.get('w')
plt.ioff()
plot_spikes(spikes)
plot_weights(weights)
plt.show()