def BindsNET_cpu(n_neurons, time):
t0 = t()
torch.set_default_tensor_type("torch.FloatTensor")
t1 = t()
network = Network()
network.add_layer(Input(n=n_neurons), name="X")
network.add_layer(LIFNodes(n=n_neurons), name="Y")
network.add_connection(
Connection(source=network.layers["X"], target=network.layers["Y"]),
source="X",
target="Y",
)
data = {"X": poisson(datum=torch.rand(n_neurons), time=time)}
network.run(inputs=data, time=time)
return t() - t0, t() - t1
network.add_layer(input_layer, name="X")
network.add_layer(conv_layer, name="Y")
network.add_connection(conv_conn, source="X", target="Y")
# Train the network.
print("Begin training.\n")
if args.tensorboard:
analyzer = TensorboardAnalyzer("logs/conv")
else:
analyzer = MatplotlibAnalyzer()
for step, batch in enumerate(tqdm(train_dataloader)):
# batch contains image, label, encoded_image since an image_encoder
# was provided
# batch["encoded_image"] is in BxTxCxHxW format
inputs = {"X": batch["encoded_image"]}
# Run the network on the input.
# Specify the location of the time dimension
network.run(inputs=inputs, time=time, input_time_dim=1)
network.reset_state_variables() # Reset state variables.
analyzer.plot_conv2d_weights(conv_conn.w, step=step)
analyzer.finalize_step()
shuffle=True,
num_workers=0,
pin_memory=gpu)
# Run training data on reservoir computer and store (spikes per neuron, label) per example.
n_iters = examples
training_pairs = []
pbar = tqdm(enumerate(dataloader))
for (i, dataPoint) in pbar:
if i > n_iters:
break
datum = dataPoint["encoded_image"].view(time, 1, 1, 28, 28).to(device_id)
label = dataPoint["label"]
pbar.set_description_str("Train progress: (%d / %d)" % (i, n_iters))
network.run(inputs={"I": datum}, time=time, input_time_dim=1)
training_pairs.append([spikes["O"].get("s").sum(0), label])
if plot:
inpt_axes, inpt_ims = plot_input(
dataPoint["image"].view(28, 28),
datum.view(time, 784).sum(0).view(28, 28),
label=label,
axes=inpt_axes,
ims=inpt_ims,
)
spike_ims, spike_axes = plot_spikes(
{layer: spikes[layer].get("s").view(-1, time)
for layer in spikes},
axes=spike_axes,