def _unconstrain_reparam(params, site):
name = site['name']
if name in params:
p = params[name]
support = site['fn'].support
if support in [real, real_vector]:
return p
t = biject_to(support)
value = t(p)
log_det = t.log_abs_det_jacobian(p, value)
log_det = sum_rightmost(
log_det,
jnp.ndim(log_det) - jnp.ndim(value) + len(site['fn'].event_shape))
if site['scale'] is not None:
log_det = site['scale'] * log_det
numpyro.factor('_{}_log_det'.format(name), log_det)
return value
def __call__(self, *args, **kwargs):
"""
An automatic guide with the same ``*args, **kwargs`` as the base ``model``.
:return: A dict mapping sample site name to sampled value.
:rtype: dict
"""
sample_latent_fn = self._sample_latent
if self.prototype_trace is None:
# run model to inspect the model structure
params = self.setup(*args, **kwargs)
sample_latent_fn = substitute(sample_latent_fn, params)
base_dist = kwargs.pop('base_dist', None)
if base_dist is None:
base_dist = _Normal(np.zeros(self.latent_size), 1.)
latent = sample_latent_fn(base_dist, *args, **kwargs)
# unpack continuous latent samples
result = {}
for name, unconstrained_value in self.unpack_latent(latent).items():
transform = self._inv_transforms[name]
site = self.prototype_trace[name]
value = transform(unconstrained_value)
log_density = -transform.log_abs_det_jacobian(
unconstrained_value, value)
if site['intermediates']:
event_ndim = len(site['fn'].base_dist.event_shape)
else:
event_ndim = len(site['fn'].event_shape)
log_density = sum_rightmost(
log_density,
np.ndim(log_density) - np.ndim(value) + event_ndim)
delta_dist = dist.Delta(value,
log_density=log_density,
event_ndim=event_ndim)
result[name] = sample(name, delta_dist)
return result
def log_prob(self, value):
log_prob = self.base_dist.log_prob(value)
return sum_rightmost(log_prob, self.reinterpreted_batch_ndims)
def log_abs_det_jacobian(self, x, y, intermediates=None):
result = self.base_transform.log_abs_det_jacobian(x, y, intermediates=intermediates)
if jnp.ndim(result) < self.reinterpreted_batch_ndims:
expected = self.domain.event_dim
raise ValueError(f"Expected x.dim() >= {expected} but got {jnp.ndim(x)}")
return sum_rightmost(result, self.reinterpreted_batch_ndims)
def log_prob(self, value):
log_prob = jnp.log(value == self.v)
log_prob = sum_rightmost(log_prob, len(self.event_shape))
return log_prob + self.log_density
def log_abs_det_jacobian(self, x, y, intermediates=None):
return sum_rightmost(
np.broadcast_to(np.log(np.abs(self.scale)), np.shape(x)),
self.event_dim)
def log_prob(self, x):
if self._validate_args:
self._validate_sample(x)
log_prob = np.log(x == self.value)
log_prob = sum_rightmost(log_prob, len(self.event_shape))
return log_prob + self.log_density