def test_normalnormal_run(self):
with self.test_session() as sess:
x_data = np.array([0.0] * 50, dtype=np.float32)
mu = Normal(mu=0.0, sigma=1.0)
x = Normal(mu=tf.ones(50) * mu, sigma=1.0)
qmu_mu = tf.Variable(tf.random_normal([]))
qmu_sigma = tf.nn.softplus(tf.Variable(tf.random_normal([])))
qmu = Normal(mu=qmu_mu, sigma=qmu_sigma)
# analytic solution: N(mu=0.0, sigma=\sqrt{1/51}=0.140)
inference = ed.KLpq({mu: qmu}, data={x: x_data})
inference.run(n_samples=25, n_iter=100)
self.assertAllClose(qmu.mean().eval(), 0, rtol=1e-1, atol=1e-1)
self.assertAllClose(qmu.std().eval(), np.sqrt(1 / 51),
rtol=1e-1, atol=1e-1)
sess = ed.get_session()
tf.global_variables_initializer().run()
i = 0
for _ in range(inference.n_iter):
X_batch, y_batch, i = next_batch(M, i)
for _ in range(5):
info_dict_d = inference.update(variables="Disc",
feed_dict={
X: X_batch,
y_ph: y_batch
})
info_dict = inference.update(variables="Gen",
feed_dict={
X: X_batch,
y_ph: y_batch
})
info_dict['loss_d'] = info_dict_d['loss_d']
info_dict['t'] = info_dict['t'] // 6 # say set of 6 updates is 1 iteration
t = info_dict['t']
inference.print_progress(info_dict)
if t == 1 or t % inference.n_print == 0:
# Check inferred posterior parameters.
mean, std = sess.run([qw.mean(), qw.std()])
print("\nInferred mean & std:")
print(mean)
print(std)
