def test_probeable():
def check_neuron_type(neuron_type, expected):
assert neuron_type.probeable == expected
ens = nengo.Ensemble(10, 1, neuron_type=neuron_type)
assert ens.neurons.probeable == expected + ("input", )
with nengo.Network():
check_neuron_type(Direct(), ("output", ))
check_neuron_type(RectifiedLinear(), ("output", ))
check_neuron_type(SpikingRectifiedLinear(), ("output", "voltage"))
check_neuron_type(Sigmoid(), ("output", ))
check_neuron_type(Tanh(), ("output", ))
check_neuron_type(LIFRate(), ("output", ))
check_neuron_type(LIF(), ("output", "voltage", "refractory_time"))
check_neuron_type(AdaptiveLIFRate(), ("output", "adaptation"))
check_neuron_type(
AdaptiveLIF(),
("output", "voltage", "refractory_time", "adaptation"))
check_neuron_type(Izhikevich(), ("output", "voltage", "recovery"))
check_neuron_type(RegularSpiking(LIFRate()),
("output", "rate_out", "voltage"))
check_neuron_type(StochasticSpiking(AdaptiveLIFRate()),
("output", "rate_out", "adaptation"))
check_neuron_type(PoissonSpiking(LIFRate()), ("output", "rate_out"))
def __init__(self, tau_ref=0.0025, amplitude=1.0, initial_state=None):
super().__init__(
base_type=Tanh(tau_ref=tau_ref),
amplitude=amplitude,
initial_state=initial_state,
)
def test_tanh_invalid():
tanh = Tanh(tau_ref=0.5)
with pytest.raises(ValidationError, match="Max rates must be below"):
tanh.gain_bias(max_rates=np.array([100]), intercepts=np.array([0]))
for max_rate, intercept in [(300.0, 1.1), (300.0, 1.0), (100.0, 0.9), (100, 1.0)]:
with nengo.Network() as net:
nengo.Ensemble(
1,
1,
neuron_type=Sigmoid(),
max_rates=[max_rate],
intercepts=[intercept],
)
with pytest.raises(BuildError, match="lead to neurons with negative"):
with Simulator(net):
pass
@pytest.mark.slow
@pytest.mark.parametrize("base_type", [LIFRate(), RectifiedLinear(), Tanh()])
def test_spiking_types(base_type, seed, plt, allclose):
spiking_types = {
RegularSpiking: dict(atol=0.05, rmse_target=0.011),
PoissonSpiking: dict(atol=0.13, rmse_target=0.024),
StochasticSpiking: dict(atol=0.10, rmse_target=0.019),
}
n_neurons = 1000
with nengo.Network(seed=seed) as net:
u = nengo.Node(lambda t: np.sin(2 * np.pi * t))
a = nengo.Ensemble(n_neurons, 1)
nengo.Connection(u, a)
u_p = nengo.Probe(u, synapse=0.005)
a_p = nengo.Probe(a, synapse=0.005)
def test_tanh_invalid():
"""Ensures validation error is thrown when given a large max rate."""
tanh = Tanh(tau_ref=0.5)
with pytest.raises(ValidationError, match="Max rates must be below"):
tanh.gain_bias(max_rates=np.array([100]), intercepts=np.array([0]))