Class RelaxedBernoulli
Inherits From: TransformedDistribution
Defined in tensorflow/contrib/distributions/python/ops/relaxed_bernoulli.py
.
RelaxedBernoulli distribution with temperature and logits parameters.
The RelaxedBernoulli is a distribution over the unit interval (0,1), which
continuously approximates a Bernoulli. The degree of approximation is
controlled by a temperature: as the temperature goes to 0 the
RelaxedBernoulli becomes discrete with a distribution described by the
logits
or probs
parameters, as the temperature goes to infinity the
RelaxedBernoulli becomes the constant distribution that is identically 0.5.
The RelaxedBernoulli distribution is a reparameterized continuous distribution that is the binary special case of the RelaxedOneHotCategorical distribution (Maddison et al., 2016; Jang et al., 2016). For details on the binary special case see the appendix of Maddison et al. (2016) where it is referred to as BinConcrete. If you use this distribution, please cite both papers.
Some care needs to be taken for loss functions that depend on the
log-probability of RelaxedBernoullis, because computing log-probabilities of
the RelaxedBernoulli can suffer from underflow issues. In many case loss
functions such as these are invariant under invertible transformations of
the random variables. The KL divergence, found in the variational autoencoder
loss, is an example. Because RelaxedBernoullis are sampled by a Logistic
random variable followed by a tf.sigmoid
op, one solution is to treat
the Logistic as the random variable and tf.sigmoid
as downstream. The
KL divergences of two Logistics, which are always followed by a tf.sigmoid
op, is equivalent to evaluating KL divergences of RelaxedBernoulli samples.
See Maddison et al., 2016 for more details where this distribution is called
the BinConcrete.
An alternative approach is to evaluate Bernoulli log probability or KL directly on relaxed samples, as done in Jang et al., 2016. In this case, guarantees on the loss are usually violated. For instance, using a Bernoulli KL in a relaxed ELBO is no longer a lower bound on the log marginal probability of the observation. Thus care and early stopping are important.
Examples
Creates three continuous distributions, which approximate 3 Bernoullis with probabilities (0.1, 0.5, 0.4). Samples from these distributions will be in the unit interval (0,1).
temperature = 0.5
p = [0.1, 0.5, 0.4]
dist = RelaxedBernoulli(temperature, probs=p)
Creates three continuous distributions, which approximate 3 Bernoullis with logits (-2, 2, 0). Samples from these distributions will be in the unit interval (0,1).
temperature = 0.5
logits = [-2, 2, 0]
dist = RelaxedBernoulli(temperature, logits=logits)
Creates three continuous distributions, whose sigmoid approximate 3 Bernoullis with logits (-2, 2, 0).
temperature = 0.5
logits = [-2, 2, 0]
dist = Logistic(logits/temperature, 1./temperature)
samples = dist.sample()
sigmoid_samples = tf.sigmoid(samples)
# sigmoid_samples has the same distribution as samples from
# RelaxedBernoulli(temperature, logits=logits)
Creates three continuous distributions, which approximate 3 Bernoullis with logits (-2, 2, 0). Samples from these distributions will be in the unit interval (0,1). Because the temperature is very low, samples from these distributions are almost discrete, usually taking values very close to 0 or 1.
temperature = 1e-5
logits = [-2, 2, 0]
dist = RelaxedBernoulli(temperature, logits=logits)
Creates three continuous distributions, which approximate 3 Bernoullis with logits (-2, 2, 0). Samples from these distributions will be in the unit interval (0,1). Because the temperature is very high, samples from these distributions are usually close to the (0.5, 0.5, 0.5) vector.
temperature = 100
logits = [-2, 2, 0]
dist = RelaxedBernoulli(temperature, logits=logits)
Chris J. Maddison, Andriy Mnih, and Yee Whye Teh. The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables. 2016.
Eric Jang, Shixiang Gu, and Ben Poole. Categorical Reparameterization with Gumbel-Softmax. 2016.
__init__
__init__(
temperature,
logits=None,
probs=None,
validate_args=False,
allow_nan_stats=True,
name='RelaxedBernoulli'
)
Construct RelaxedBernoulli distributions. (deprecated)
Args:
temperature
: An 0-DTensor
, representing the temperature of a set of RelaxedBernoulli distributions. The temperature should be positive.logits
: An N-DTensor
representing the log-odds of a positive event. Each entry in theTensor
parametrizes an independent RelaxedBernoulli distribution where the probability of an event is sigmoid(logits). Only one oflogits
orprobs
should be passed in.probs
: An N-DTensor
representing the probability of a positive event. Each entry in theTensor
parameterizes an independent Bernoulli distribution. Only one oflogits
orprobs
should be passed in.validate_args
: Pythonbool
, defaultFalse
. WhenTrue
distribution parameters are checked for validity despite possibly degrading runtime performance. WhenFalse
invalid inputs may silently render incorrect outputs.allow_nan_stats
: Pythonbool
, defaultTrue
. WhenTrue
, statistics (e.g., mean, mode, variance) use the value "NaN
" to indicate the result is undefined. WhenFalse
, an exception is raised if one or more of the statistic's batch members are undefined.name
: Pythonstr
name prefixed to Ops created by this class.
Raises:
ValueError
: If bothprobs
andlogits
are passed, or if neither.
Properties
allow_nan_stats
Python bool
describing behavior when a stat is undefined.
Stats return +/- infinity when it makes sense. E.g., the variance of a Cauchy distribution is infinity. However, sometimes the statistic is undefined, e.g., if a distribution's pdf does not achieve a maximum within the support of the distribution, the mode is undefined. If the mean is undefined, then by definition the variance is undefined. E.g. the mean for Student's T for df = 1 is undefined (no clear way to say it is either + or - infinity), so the variance = E[(X - mean)**2] is also undefined.
Returns:
allow_nan_stats
: Pythonbool
.
batch_shape
Shape of a single sample from a single event index as a TensorShape
.
May be partially defined or unknown.
The batch dimensions are indexes into independent, non-identical parameterizations of this distribution.
Returns:
batch_shape
:TensorShape
, possibly unknown.
bijector
Function transforming x => y.
distribution
Base distribution, p(x).
dtype
The DType
of Tensor
s handled by this Distribution
.
event_shape
Shape of a single sample from a single batch as a TensorShape
.
May be partially defined or unknown.
Returns:
event_shape
:TensorShape
, possibly unknown.
logits
Log-odds of 1
.
name
Name prepended to all ops created by this Distribution
.
parameters
Dictionary of parameters used to instantiate this Distribution
.
probs
Probability of 1
.
reparameterization_type
Describes how samples from the distribution are reparameterized.
Currently this is one of the static instances
distributions.FULLY_REPARAMETERIZED
or distributions.NOT_REPARAMETERIZED
.
Returns:
An instance of ReparameterizationType
.
temperature
Distribution parameter for the location.
validate_args
Python bool
indicating possibly expensive checks are enabled.
Methods
tf.contrib.distributions.RelaxedBernoulli.batch_shape_tensor
batch_shape_tensor(name='batch_shape_tensor')
Shape of a single sample from a single event index as a 1-D Tensor
.
The batch dimensions are indexes into independent, non-identical parameterizations of this distribution.
Args:
name
: name to give to the op
Returns:
batch_shape
:Tensor
.
tf.contrib.distributions.RelaxedBernoulli.cdf
cdf(
value,
name='cdf'
)
Cumulative distribution function.
Given random variable X
, the cumulative distribution function cdf
is:
cdf(x) := P[X <= x]
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
cdf
: aTensor
of shapesample_shape(x) + self.batch_shape
with values of typeself.dtype
.
tf.contrib.distributions.RelaxedBernoulli.copy
copy(**override_parameters_kwargs)
Creates a deep copy of the distribution.
Args:
**override_parameters_kwargs
: String/value dictionary of initialization arguments to override with new values.
Returns:
distribution
: A new instance oftype(self)
initialized from the union of self.parameters and override_parameters_kwargs, i.e.,dict(self.parameters, **override_parameters_kwargs)
.
tf.contrib.distributions.RelaxedBernoulli.covariance
covariance(name='covariance')
Covariance.
Covariance is (possibly) defined only for non-scalar-event distributions.
For example, for a length-k
, vector-valued distribution, it is calculated
as,
Cov[i, j] = Covariance(X_i, X_j) = E[(X_i - E[X_i]) (X_j - E[X_j])]
where Cov
is a (batch of) k x k
matrix, 0 <= (i, j) < k
, and E
denotes expectation.
Alternatively, for non-vector, multivariate distributions (e.g.,
matrix-valued, Wishart), Covariance
shall return a (batch of) matrices
under some vectorization of the events, i.e.,
Cov[i, j] = Covariance(Vec(X)_i, Vec(X)_j) = [as above]
where Cov
is a (batch of) k' x k'
matrices,
0 <= (i, j) < k' = reduce_prod(event_shape)
, and Vec
is some function
mapping indices of this distribution's event dimensions to indices of a
length-k'
vector.
Args:
name
: Pythonstr
prepended to names of ops created by this function.
Returns:
covariance
: Floating-pointTensor
with shape[B1, ..., Bn, k', k']
where the firstn
dimensions are batch coordinates andk' = reduce_prod(self.event_shape)
.
tf.contrib.distributions.RelaxedBernoulli.cross_entropy
cross_entropy(
other,
name='cross_entropy'
)
Computes the (Shannon) cross entropy.
Denote this distribution (self
) by P
and the other
distribution by
Q
. Assuming P, Q
are absolutely continuous with respect to
one another and permit densities p(x) dr(x)
and q(x) dr(x)
, (Shanon)
cross entropy is defined as:
H[P, Q] = E_p[-log q(X)] = -int_F p(x) log q(x) dr(x)
where F
denotes the support of the random variable X ~ P
.
Args:
other
:tfp.distributions.Distribution
instance.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
cross_entropy
:self.dtype
Tensor
with shape[B1, ..., Bn]
representingn
different calculations of (Shanon) cross entropy.
tf.contrib.distributions.RelaxedBernoulli.entropy
entropy(name='entropy')
Shannon entropy in nats.
tf.contrib.distributions.RelaxedBernoulli.event_shape_tensor
event_shape_tensor(name='event_shape_tensor')
Shape of a single sample from a single batch as a 1-D int32 Tensor
.
Args:
name
: name to give to the op
Returns:
event_shape
:Tensor
.
tf.contrib.distributions.RelaxedBernoulli.is_scalar_batch
is_scalar_batch(name='is_scalar_batch')
Indicates that batch_shape == []
.
Args:
name
: Pythonstr
prepended to names of ops created by this function.
Returns:
is_scalar_batch
:bool
scalarTensor
.
tf.contrib.distributions.RelaxedBernoulli.is_scalar_event
is_scalar_event(name='is_scalar_event')
Indicates that event_shape == []
.
Args:
name
: Pythonstr
prepended to names of ops created by this function.
Returns:
is_scalar_event
:bool
scalarTensor
.
tf.contrib.distributions.RelaxedBernoulli.kl_divergence
kl_divergence(
other,
name='kl_divergence'
)
Computes the Kullback--Leibler divergence.
Denote this distribution (self
) by p
and the other
distribution by
q
. Assuming p, q
are absolutely continuous with respect to reference
measure r
, the KL divergence is defined as:
KL[p, q] = E_p[log(p(X)/q(X))]
= -int_F p(x) log q(x) dr(x) + int_F p(x) log p(x) dr(x)
= H[p, q] - H[p]
where F
denotes the support of the random variable X ~ p
, H[., .]
denotes (Shanon) cross entropy, and H[.]
denotes (Shanon) entropy.
Args:
other
:tfp.distributions.Distribution
instance.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
kl_divergence
:self.dtype
Tensor
with shape[B1, ..., Bn]
representingn
different calculations of the Kullback-Leibler divergence.
tf.contrib.distributions.RelaxedBernoulli.log_cdf
log_cdf(
value,
name='log_cdf'
)
Log cumulative distribution function.
Given random variable X
, the cumulative distribution function cdf
is:
log_cdf(x) := Log[ P[X <= x] ]
Often, a numerical approximation can be used for log_cdf(x)
that yields
a more accurate answer than simply taking the logarithm of the cdf
when
x << -1
.
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
logcdf
: aTensor
of shapesample_shape(x) + self.batch_shape
with values of typeself.dtype
.
tf.contrib.distributions.RelaxedBernoulli.log_prob
log_prob(
value,
name='log_prob'
)
Log probability density/mass function.
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
log_prob
: aTensor
of shapesample_shape(x) + self.batch_shape
with values of typeself.dtype
.
tf.contrib.distributions.RelaxedBernoulli.log_survival_function
log_survival_function(
value,
name='log_survival_function'
)
Log survival function.
Given random variable X
, the survival function is defined:
log_survival_function(x) = Log[ P[X > x] ]
= Log[ 1 - P[X <= x] ]
= Log[ 1 - cdf(x) ]
Typically, different numerical approximations can be used for the log
survival function, which are more accurate than 1 - cdf(x)
when x >> 1
.
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
Tensor
of shape sample_shape(x) + self.batch_shape
with values of type
self.dtype
.
tf.contrib.distributions.RelaxedBernoulli.mean
mean(name='mean')
Mean.
tf.contrib.distributions.RelaxedBernoulli.mode
mode(name='mode')
Mode.
tf.contrib.distributions.RelaxedBernoulli.param_shapes
param_shapes(
cls,
sample_shape,
name='DistributionParamShapes'
)
Shapes of parameters given the desired shape of a call to sample()
.
This is a class method that describes what key/value arguments are required
to instantiate the given Distribution
so that a particular shape is
returned for that instance's call to sample()
.
Subclasses should override class method _param_shapes
.
Args:
sample_shape
:Tensor
or python list/tuple. Desired shape of a call tosample()
.name
: name to prepend ops with.
Returns:
dict
of parameter name to Tensor
shapes.
tf.contrib.distributions.RelaxedBernoulli.param_static_shapes
param_static_shapes(
cls,
sample_shape
)
param_shapes with static (i.e. TensorShape
) shapes.
This is a class method that describes what key/value arguments are required
to instantiate the given Distribution
so that a particular shape is
returned for that instance's call to sample()
. Assumes that the sample's
shape is known statically.
Subclasses should override class method _param_shapes
to return
constant-valued tensors when constant values are fed.
Args:
sample_shape
:TensorShape
or python list/tuple. Desired shape of a call tosample()
.
Returns:
dict
of parameter name to TensorShape
.
Raises:
ValueError
: ifsample_shape
is aTensorShape
and is not fully defined.
tf.contrib.distributions.RelaxedBernoulli.prob
prob(
value,
name='prob'
)
Probability density/mass function.
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
prob
: aTensor
of shapesample_shape(x) + self.batch_shape
with values of typeself.dtype
.
tf.contrib.distributions.RelaxedBernoulli.quantile
quantile(
value,
name='quantile'
)
Quantile function. Aka "inverse cdf" or "percent point function".
Given random variable X
and p in [0, 1]
, the quantile
is:
quantile(p) := x such that P[X <= x] == p
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
quantile
: aTensor
of shapesample_shape(x) + self.batch_shape
with values of typeself.dtype
.
tf.contrib.distributions.RelaxedBernoulli.sample
sample(
sample_shape=(),
seed=None,
name='sample'
)
Generate samples of the specified shape.
Note that a call to sample()
without arguments will generate a single
sample.
Args:
sample_shape
: 0D or 1Dint32
Tensor
. Shape of the generated samples.seed
: Python integer seed for RNGname
: name to give to the op.
Returns:
samples
: aTensor
with prepended dimensionssample_shape
.
tf.contrib.distributions.RelaxedBernoulli.stddev
stddev(name='stddev')
Standard deviation.
Standard deviation is defined as,
stddev = E[(X - E[X])**2]**0.5
where X
is the random variable associated with this distribution, E
denotes expectation, and stddev.shape = batch_shape + event_shape
.
Args:
name
: Pythonstr
prepended to names of ops created by this function.
Returns:
stddev
: Floating-pointTensor
with shape identical tobatch_shape + event_shape
, i.e., the same shape asself.mean()
.
tf.contrib.distributions.RelaxedBernoulli.survival_function
survival_function(
value,
name='survival_function'
)
Survival function.
Given random variable X
, the survival function is defined:
survival_function(x) = P[X > x]
= 1 - P[X <= x]
= 1 - cdf(x).
Args:
value
:float
ordouble
Tensor
.name
: Pythonstr
prepended to names of ops created by this function.
Returns:
Tensor
of shape sample_shape(x) + self.batch_shape
with values of type
self.dtype
.
tf.contrib.distributions.RelaxedBernoulli.variance
variance(name='variance')
Variance.
Variance is defined as,
Var = E[(X - E[X])**2]
where X
is the random variable associated with this distribution, E
denotes expectation, and Var.shape = batch_shape + event_shape
.
Args:
name
: Pythonstr
prepended to names of ops created by this function.
Returns:
variance
: Floating-pointTensor
with shape identical tobatch_shape + event_shape
, i.e., the same shape asself.mean()
.