TransientDist

class liesel.model.nodes.TransientDist(distribution, *inputs, _name='', _needs_seed=False, **kwinputs)

Bases: TransientNode, Dist

A transient distribution node that does not cache its value.

Methods

add_inputs(*inputs, **kwinputs)	Adds non-keyword and keyword input nodes to the existing ones.
all_input_nodes()	Returns all non-keyword and keyword input nodes as a unique tuple.
all_output_nodes()	Returns all output nodes as a unique tuple.
clear_state()	Clears the state of the node.
flag_outdated()	Flags the node and its recursive outputs as outdated.
init_dist()	Initializes the distribution.
set_inputs(*inputs, **kwinputs)	Sets the non-keyword and keyword input nodes.
update()	Does nothing.

Attributes

at	Where to evaluate the distribution.
distribution	The wrapped distribution.
groups	The groups that this node is a part of.
inputs	The non-keyword input nodes.
kwinputs	The keyword input nodes.
log_prob	The log-probability of the distribution.
model	The model the node is part of.
name	The name of the node.
needs_seed	Whether the node needs a seed / PRNG key.
outdated	Whether the node is outdated.
outputs	The output nodes.
per_obs	Whether the log-probability is stored per observation or summed up.
state	The state of the node with the value None.
value	The value of the node.
var	The variable the node is part of.
monitor	Whether the node should be monitored by an inference algorithm.