def testBrownianMotionMarkovChainLogprobMatchesOriginal(self):
model = (brownian_motion.BrownianMotionMissingMiddleObservations(
use_markov_chain=False))
markov_chain_model = (
brownian_motion.BrownianMotionMissingMiddleObservations(
use_markov_chain=True))
x = self.evaluate(model.prior_distribution().sample(
400, seed=tfp_test_util.test_seed()))
self.assertAllClose(model.unnormalized_log_prob(x),
markov_chain_model.unnormalized_log_prob(
tf.stack(x, axis=-1)),
atol=1e-2)
def testBrownianMotionMissingMiddleObservations(self):
"""Checks that unconstrained parameters yield finite joint densities."""
model = brownian_motion.BrownianMotionMissingMiddleObservations()
self.validate_log_prob_and_transforms(
model,
sample_transformation_shapes=dict(identity=[30]),
check_ground_truth_mean_standard_error=True,
check_ground_truth_mean=True,
check_ground_truth_standard_deviation=True)
def testBrownianMotionMissingMiddleObservationsHMC(self):
"""Checks approximate samples from the model against the ground truth."""
model = brownian_motion.BrownianMotionMissingMiddleObservations()
self.validate_ground_truth_using_hmc(
model,
num_chains=4,
num_steps=4000,
num_leapfrog_steps=18,
step_size=0.02,
)