def __init__(self, logits, dtype=None, group_event_ndims=0):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype(
[(self._logits, 'Categorical.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
static_logits_shape = self._logits.get_shape()
shape_err_msg = "logits should have rank >= 1."
if static_logits_shape and (static_logits_shape.ndims < 1):
raise ValueError(shape_err_msg)
elif static_logits_shape and (
static_logits_shape[-1].value is not None):
self._n_categories = static_logits_shape[-1].value
else:
_assert_shape_op = tf.assert_rank_at_least(
self._logits, 1, message=shape_err_msg)
with tf.control_dependencies([_assert_shape_op]):
self._logits = tf.identity(self._logits)
self._n_categories = tf.shape(self._logits)[-1]
super(Categorical, self).__init__(
dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_event_ndims=group_event_ndims)
def __init__(self,
logits,
n_experiments,
dtype=None,
group_ndims=0,
**kwargs):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype([(self._logits,
'Multinomial.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
self._logits, self._n_categories = assert_rank_at_least_one(
self._logits, 'Multinomial.logits')
self._n_experiments = assert_positive_int32_integer(
n_experiments, 'Multinomial.n_experiments')
super(Multinomial, self).__init__(dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_ndims=group_ndims,
**kwargs)
def __init__(self,
logits,
normalize_logits=True,
dtype=None,
group_ndims=0,
**kwargs):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype(
[(self._logits, 'UnnormalizedMultinomial.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
self._logits = assert_rank_at_least_one(
self._logits, 'UnnormalizedMultinomial.logits')
self._n_categories = get_shape_at(self._logits, -1)
self.normalize_logits = normalize_logits
super(UnnormalizedMultinomial, self).__init__(
dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_ndims=group_ndims,
**kwargs)
def __init__(self,
logits,
n_experiments,
dtype=None,
group_event_ndims=0):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype(
[(self._logits, 'Multinomial.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
static_logits_shape = self._logits.get_shape()
shape_err_msg = "logits should have rank >= 1."
if static_logits_shape and (static_logits_shape.ndims < 1):
raise ValueError(shape_err_msg)
elif static_logits_shape and (
static_logits_shape[-1].value is not None):
self._n_categories = static_logits_shape[-1].value
else:
_assert_shape_op = tf.assert_rank_at_least(
self._logits, 1, message=shape_err_msg)
with tf.control_dependencies([_assert_shape_op]):
self._logits = tf.identity(self._logits)
self._n_categories = tf.shape(self._logits)[-1]
sign_err_msg = "n_experiments must be positive"
if isinstance(n_experiments, int):
if n_experiments <= 0:
raise ValueError(sign_err_msg)
self._n_experiments = n_experiments
else:
try:
n_experiments = tf.convert_to_tensor(n_experiments, tf.int32)
except ValueError:
raise TypeError('n_experiments must be int32')
_assert_rank_op = tf.assert_rank(
n_experiments, 0,
message="n_experiments should be a scalar (0-D Tensor).")
_assert_positive_op = tf.assert_greater(
n_experiments, 0, message=sign_err_msg)
with tf.control_dependencies([_assert_rank_op,
_assert_positive_op]):
self._n_experiments = tf.identity(n_experiments)
super(Multinomial, self).__init__(
dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_event_ndims=group_event_ndims)
def __init__(self, logits, dtype=None, group_event_ndims=0):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype([(self._logits,
'Bernoulli.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
super(Bernoulli, self).__init__(dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_event_ndims=group_event_ndims)
def __init__(self, logits, dtype=None, group_event_ndims=0):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype([(self._logits,
'Categorical.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
self._logits, self._n_categories = assert_rank_at_least_one(
self._logits, 'Categorical.logits')
static_logits_shape = self._logits.get_shape()
super(Categorical, self).__init__(dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_event_ndims=group_event_ndims)
def __init__(self,
rate,
dtype=None,
group_event_ndims=0,
check_numerics=False):
self._rate = tf.convert_to_tensor(rate)
param_dtype = assert_same_float_dtype([(self._rate, 'Poisson.rate')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
self._check_numerics = check_numerics
super(Poisson, self).__init__(dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_event_ndims=group_event_ndims)
def __init__(self,
logits,
n_experiments,
dtype=None,
group_ndims=0,
check_numerics=False,
**kwargs):
self._logits = tf.convert_to_tensor(logits)
param_dtype = assert_same_float_dtype([(self._logits,
'Binomial.logits')])
if dtype is None:
dtype = tf.int32
assert_same_float_and_int_dtype([], dtype)
sign_err_msg = "n_experiments must be positive"
if isinstance(n_experiments, int):
if n_experiments <= 0:
raise ValueError(sign_err_msg)
self._n_experiments = n_experiments
else:
try:
n_experiments = tf.convert_to_tensor(n_experiments, tf.int32)
except ValueError:
raise TypeError('n_experiments must be int32')
_assert_rank_op = tf.assert_rank(
n_experiments,
0,
message="n_experiments should be a scalar (0-D Tensor).")
_assert_positive_op = tf.assert_greater(n_experiments,
0,
message=sign_err_msg)
with tf.control_dependencies(
[_assert_rank_op, _assert_positive_op]):
self._n_experiments = tf.identity(n_experiments)
self._check_numerics = check_numerics
super(Binomial, self).__init__(dtype=dtype,
param_dtype=param_dtype,
is_continuous=False,
is_reparameterized=False,
group_ndims=group_ndims,
**kwargs)
