def get_h_and_v(self, num_samples):
z = sample_from_bernoulli(
tf.constant(.5,
shape=(num_samples, self.dim_z2h[0]),
dtype=tf.float32))
h = gumbel_sigmoid_sample(self.z2h(z), self.temp, self.hard)
v = gumbel_sigmoid_sample(self.h2v(h), self.temp, self.hard)
return h, v
def sample(self, num_samples):
z = sample_from_bernoulli(
tf.constant(.5,
shape=(num_samples, self.dims[0]),
dtype=tf.float32))
h2 = gumbel_sigmoid_sample(self.inference(z), self.temp, self.hard)
return h2
def logprobs_all(self, num_samples):
z = sample_from_bernoulli(
tf.constant(.5,
shape=(num_samples, self.dims[0]),
dtype=tf.float32))
mu = self.inference(z)
h = gumbel_sigmoid_sample(mu, self.temp, self.hard)
logp_z = bernoulli_log_likelihood(
z, tf.constant(.5, shape=z.shape, dtype=tf.float32))
logp_h_given_z = bernoulli_log_likelihood(h, mu)
return logp_z + logp_h_given_z, z, h
def _test_gumbel(self, num_samples=1):
z = sample_from_bernoulli(
tf.constant(.5,
shape=(num_samples, self.dim_z2h[0]),
dtype=tf.float32))
h_mu = self.z2h(z)
h_hard = gumbel_sigmoid_sample(h_mu, self.temp, True)
h_soft = gumbel_sigmoid_sample(h_mu, self.temp, False)
print 'h_mu\n', h_mu, 'h_hard\n', h_hard, 'h_soft\n', h_soft
v_mu = self.h2v(h_hard)
v_hard = gumbel_sigmoid_sample(v_mu, self.temp, True)
v_soft = gumbel_sigmoid_sample(v_mu, self.temp, False)
print 'v_mu\n', v_mu, 'v_hard\n', v_hard, 'v_soft\n', v_soft
v_mu = self.h2v(h_soft)
v_hard = gumbel_sigmoid_sample(v_mu, self.temp, True)
v_soft = gumbel_sigmoid_sample(v_mu, self.temp, False)
print 'v_mu\n', v_mu, 'v_hard\n', v_hard, 'v_soft\n', v_soft
exit()
def log_conditional_prob(self, x):
# E_{Q(h|x)} log Q(h|x)
h_mu = self.inference(x)
h = gumbel_sigmoid_sample(h_mu, self.temp, self.hard)
logp_h_given_x = bernoulli_log_likelihood(h, h_mu)
return logp_h_given_x, h
def get_h_soft(self, x):
return gumbel_sigmoid_sample(self.inference(x), self.temp, self.hard)
def get_h_and_v(self, num_samples):
h = self.prior_h.sample(num_samples)
v = gumbel_sigmoid_sample(self.h2v(h), self.temp, self.hard)
return h, v
def sample(self, num_samples):
index = gumbel_softmax_sample(
tf.tile(self.pi_logits, [num_samples, 1]), self.temp)
sample_mus = tf.matmul(index, self.mus)
return gumbel_sigmoid_sample(sample_mus, self.temp, self.hard)
