def test_dist_poisson_multivariate_batched(self):
dist_sample_shape_correct = [2, 3]
dist_means_correct = [[1, 2, 15], [100, 200, 300]]
dist_stddevs_correct = [[math.sqrt(1), math.sqrt(2), math.sqrt(15)], [math.sqrt(100), math.sqrt(200), math.sqrt(300)]]
dist_rates_correct = [[1, 2, 15], [100, 200, 300]]
dist_log_probs_correct = [[sum([-1, -1.30685, -2.27852])], [sum([-3.22236, -3.56851, -3.77110])]]
dist = Poisson(dist_rates_correct)
dist_sample_shape = list(dist.sample().size())
dist_empirical = Empirical([dist.sample() for i in range(empirical_samples)])
dist_rates = util.to_numpy(dist.rate)
dist_means = util.to_numpy(dist.mean)
dist_means_empirical = util.to_numpy(dist_empirical.mean)
dist_stddevs = util.to_numpy(dist.stddev)
dist_stddevs_empirical = util.to_numpy(dist_empirical.stddev)
dist_log_probs = util.to_numpy(dist.log_prob(dist_means_correct))
util.debug('dist_sample_shape', 'dist_sample_shape_correct', 'dist_rates', 'dist_rates_correct', 'dist_means', 'dist_means_empirical', 'dist_means_correct', 'dist_stddevs', 'dist_stddevs_empirical', 'dist_stddevs_correct', 'dist_log_probs', 'dist_log_probs_correct')
self.assertEqual(dist_sample_shape, dist_sample_shape_correct)
self.assertTrue(np.allclose(dist_means, dist_means_correct, atol=0.1))
self.assertTrue(np.allclose(dist_means_empirical, dist_means_correct, atol=0.25))
self.assertTrue(np.allclose(dist_stddevs, dist_stddevs_correct, atol=0.1))
self.assertTrue(np.allclose(dist_stddevs_empirical, dist_stddevs_correct, atol=0.25))
self.assertTrue(np.allclose(dist_rates, dist_rates_correct, atol=0.1))
self.assertTrue(np.allclose(dist_log_probs, dist_log_probs_correct, atol=0.1))
def test_dist_poisson_batched(self):
dist_sample_shape_correct = [2, 1]
dist_means_correct = [[4], [100]]
dist_stddevs_correct = [[math.sqrt(4)], [math.sqrt(100)]]
dist_rates_correct = [[4], [100]]
dist_log_probs_correct = [[-1.63288], [-3.22236]]
dist = Poisson(dist_rates_correct)
dist_sample_shape = list(dist.sample().size())
dist_empirical = Empirical([dist.sample() for i in range(empirical_samples)])
dist_rates = util.to_numpy(dist.rate)
dist_means = util.to_numpy(dist.mean)
dist_means_empirical = util.to_numpy(dist_empirical.mean)
dist_stddevs = util.to_numpy(dist.stddev)
dist_stddevs_empirical = util.to_numpy(dist_empirical.stddev)
dist_log_probs = util.to_numpy(dist.log_prob(dist_means_correct))
util.debug('dist_sample_shape', 'dist_sample_shape_correct', 'dist_rates', 'dist_rates_correct', 'dist_means', 'dist_means_empirical', 'dist_means_correct', 'dist_stddevs', 'dist_stddevs_empirical', 'dist_stddevs_correct', 'dist_log_probs', 'dist_log_probs_correct')
self.assertEqual(dist_sample_shape, dist_sample_shape_correct)
self.assertTrue(np.allclose(dist_means, dist_means_correct, atol=0.1))
self.assertTrue(np.allclose(dist_means_empirical, dist_means_correct, atol=0.1))
self.assertTrue(np.allclose(dist_stddevs, dist_stddevs_correct, atol=0.1))
self.assertTrue(np.allclose(dist_stddevs_empirical, dist_stddevs_correct, atol=0.1))
self.assertTrue(np.allclose(dist_rates, dist_rates_correct, atol=0.1))
self.assertTrue(np.allclose(dist_log_probs, dist_log_probs_correct, atol=0.1))
def test_dist_poisson_multivariate_from_flat_params(self):
dist_sample_shape_correct = [1, 3]
dist_means_correct = [[1, 2, 15]]
dist_stddevs_correct = [[math.sqrt(1), math.sqrt(2), math.sqrt(15)]]
dist_rates_correct = [[1, 2, 15]]
dist_log_probs_correct = [sum([-1, -1.30685, -2.27852])]
dist = Poisson(dist_rates_correct[0])
dist_sample_shape = list(dist.sample().size())
dist_empirical = Empirical([dist.sample() for i in range(empirical_samples)])
dist_rates = util.to_numpy(dist.rate)
dist_means = util.to_numpy(dist.mean)
dist_means_empirical = util.to_numpy(dist_empirical.mean)
dist_stddevs = util.to_numpy(dist.stddev)
dist_stddevs_empirical = util.to_numpy(dist_empirical.stddev)
dist_log_probs = util.to_numpy(dist.log_prob(dist_means_correct))
util.debug('dist_sample_shape', 'dist_sample_shape_correct', 'dist_rates', 'dist_rates_correct', 'dist_means', 'dist_means_empirical', 'dist_means_correct', 'dist_stddevs', 'dist_stddevs_empirical', 'dist_stddevs_correct', 'dist_log_probs', 'dist_log_probs_correct')
self.assertEqual(dist_sample_shape, dist_sample_shape_correct)
self.assertTrue(np.allclose(dist_means, dist_means_correct, atol=0.1))
self.assertTrue(np.allclose(dist_means_empirical, dist_means_correct, atol=0.1))
self.assertTrue(np.allclose(dist_stddevs, dist_stddevs_correct, atol=0.1))
self.assertTrue(np.allclose(dist_stddevs_empirical, dist_stddevs_correct, atol=0.1))
self.assertTrue(np.allclose(dist_rates, dist_rates_correct, atol=0.1))
self.assertTrue(np.allclose(dist_log_probs, dist_log_probs_correct, atol=0.1))
def true_posterior(self, observe=6):
count_prior = Poisson(4)
vals = []
log_weights = []
for r in range(40):
for s in range(40):
if 4 < float(r):
l = 6
else:
f = self.fibonacci(3 * r)
l = 1 + f + count_prior.sample()
vals.append(r)
log_weights.append(Poisson(l).log_prob(observe) + count_prior.log_prob(r) + count_prior.log_prob(s))
return Empirical(vals, log_weights)