def test_vonmises_logpdf(T=100, K=4, D=10):
# Test single datapoint log pdf
x = npr.vonmises(0, 1, size=(T, D))
mus = npr.randn(K, D)
kappas = np.exp(npr.randn(K, D))
ll1 = vonmises_logpdf(x[:, None, :], mus, kappas)
ll2 = np.sum(vonmises.logpdf(x[:, None, :],
kappas[None, :, :],
loc=mus[None, :, :]),
axis=-1)
assert np.allclose(ll1, ll2)
def sample_x(self, z, xhist, input=None, tag=None, with_noise=True):
D, mus, kappas = self.D, self.mus, np.exp(self.log_kappas)
assert self.D == 1, "InputDrivenObservations written for D = 1!"
if input.ndim == 1 and input.shape == (
self.M,
): # if input is vector of size self.M (one time point), expand dims to be (1, M)
input = np.expand_dims(input, axis=0)
###
driven_angle = mus @ input.T
kappas_t = np.repeat(kappas[:, :, None], input.T.shape[-1],
axis=2) #time-independent
return npr.vonmises(driven_angle[z], kappas_t[z]) #, D)
def test_vonmises_logpdf(T=100, K=4, D=10):
"""
NOTE: Skipping this test until the scipy special functions
make it into the release version of autograd.
"""
# Test single datapoint log pdf
x = npr.vonmises(0, 1, size=(T, D))
mus = npr.randn(K, D)
kappas = np.exp(npr.randn(K, D))
ll1 = vonmises_logpdf(x[:, None, :], mus, kappas)
ll2 = np.sum(vonmises.logpdf(x[:, None, :],
kappas[None, :, :],
loc=mus[None, :, :]),
axis=-1)
assert np.allclose(ll1, ll2)
def sample_x(self, z, xhist, input=None, tag=None, with_noise=True):
D, mus, kappas = self.D, self.mus, np.exp(self.log_kappas)
return npr.vonmises(self.mus[z], kappas[z], D)