def testBijector(self):
with self.test_session():
self.assertEqual("sigmoid", Sigmoid().name)
x = np.linspace(-10., 10., 100).reshape([2, 5, 10]).astype(np.float32)
y = special.expit(x)
ildj = -np.log(y) - np.log1p(-y)
self.assertAllClose(y, Sigmoid().forward(x).eval(), atol=0., rtol=1e-2)
self.assertAllClose(x, Sigmoid().inverse(y).eval(), atol=0., rtol=1e-4)
self.assertAllClose(ildj, Sigmoid().inverse_log_det_jacobian(y).eval(),
atol=0., rtol=1e-6)
self.assertAllClose(-ildj, Sigmoid().forward_log_det_jacobian(x).eval(),
atol=0., rtol=1e-4)
def testBijector(self):
with self.cached_session():
self.assertEqual("sigmoid", Sigmoid().name)
x = np.linspace(-10., 10., 100).reshape([2, 5, 10]).astype(np.float32)
y = special.expit(x)
ildj = -np.log(y) - np.log1p(-y)
bijector = Sigmoid()
self.assertAllClose(y, bijector.forward(x).eval(), atol=0., rtol=1e-2)
self.assertAllClose(x, bijector.inverse(y).eval(), atol=0., rtol=1e-4)
self.assertAllClose(ildj, bijector.inverse_log_det_jacobian(
y, event_ndims=0).eval(), atol=0., rtol=1e-6)
self.assertAllClose(-ildj, bijector.forward_log_det_jacobian(
x, event_ndims=0).eval(), atol=0., rtol=1e-4)
def testBijectiveAndFinite(self):
with self.cached_session():
x = np.linspace(-7., 7., 100).astype(np.float32)
eps = 1e-3
y = np.linspace(eps, 1. - eps, 100).astype(np.float32)
assert_bijective_and_finite(Sigmoid(),
x,
y,
event_ndims=0,
atol=0.,
rtol=1e-4)
def __init__(self,
temperature,
logits=None,
probs=None,
validate_args=False,
allow_nan_stats=True,
name="RelaxedBernoulli"):
"""Construct RelaxedBernoulli distributions.
Args:
temperature: An 0-D `Tensor`, representing the temperature
of a set of RelaxedBernoulli distributions. The temperature should be
positive.
logits: An N-D `Tensor` representing the log-odds
of a positive event. Each entry in the `Tensor` parametrizes
an independent RelaxedBernoulli distribution where the probability of an
event is sigmoid(logits). Only one of `logits` or `probs` should be
passed in.
probs: An N-D `Tensor` representing the probability of a positive event.
Each entry in the `Tensor` parameterizes an independent Bernoulli
distribution. Only one of `logits` or `probs` should be passed in.
validate_args: Python `bool`, default `False`. When `True` distribution
parameters are checked for validity despite possibly degrading runtime
performance. When `False` invalid inputs may silently render incorrect
outputs.
allow_nan_stats: Python `bool`, default `True`. When `True`, statistics
(e.g., mean, mode, variance) use the value "`NaN`" to indicate the
result is undefined. When `False`, an exception is raised if one or
more of the statistic's batch members are undefined.
name: Python `str` name prefixed to Ops created by this class.
Raises:
ValueError: If both `probs` and `logits` are passed, or if neither.
"""
parameters = dict(locals())
with ops.name_scope(name, values=[logits, probs, temperature]) as name:
with ops.control_dependencies(
[check_ops.assert_positive(temperature
)] if validate_args else []):
self._temperature = array_ops.identity(temperature,
name="temperature")
self._logits, self._probs = distribution_util.get_logits_and_probs(
logits=logits, probs=probs, validate_args=validate_args)
super(RelaxedBernoulli,
self).__init__(distribution=logistic.Logistic(
self._logits / self._temperature,
1. / self._temperature,
validate_args=validate_args,
allow_nan_stats=allow_nan_stats,
name=name + "/Logistic"),
bijector=Sigmoid(validate_args=validate_args),
validate_args=validate_args,
name=name)
self._parameters = parameters
def testScalarCongruency(self):
with self.cached_session():
assert_scalar_congruency(Sigmoid(), lower_x=-7., upper_x=7.)