def multinomial(self,
name,
logits,
n_experiments,
normalize_logits=True,
n_samples=None,
group_ndims=0,
dtype=tf.int32,
**kwargs):
"""
Add a stochastic node in this :class:`BayesianNet` that follows the
Multinomial distribution.
:param name: The name of the stochastic node. Must be unique in a
:class:`BayesianNet`.
See
:class:`~zhusuan.distributions.multivariate.Multinomial`
for more information about the other arguments.
:return: A :class:`StochasticTensor` instance.
"""
dist = distributions.Multinomial(logits,
n_experiments,
normalize_logits=normalize_logits,
group_ndims=group_ndims,
dtype=dtype,
**kwargs)
return self.stochastic(name, dist, n_samples=n_samples, **kwargs)
def __init__(self,
name,
logits,
n_experiments,
n_samples=None,
group_event_ndims=0,
dtype=None):
multinomial = distributions.Multinomial(
logits,
n_experiments,
group_event_ndims=group_event_ndims,
dtype=dtype,
)
super(Multinomial, self).__init__(name, multinomial, n_samples)
def __init__(self,
name,
logits,
n_experiments,
normalize_logits=True,
n_samples=None,
group_ndims=0,
dtype=None,
**kwargs):
multinomial = distributions.Multinomial(
logits,
n_experiments,
normalize_logits=normalize_logits,
group_ndims=group_ndims,
dtype=dtype,
**kwargs)
super(Multinomial, self).__init__(name, multinomial, n_samples)
