Model.simulate()#
- Model.simulate(seed, skip=())[source]#
Updates the model state simulating from the probability distributions in the model using a provided random seed, optionally skipping specified nodes.
- Parameters:
seed – The seed is split and distributed to the distribution nodes in the model. Must be a
KeyArray
, i.e. an array of shape (2,) and dtypeuint32
. Seejax.random
for more details.skip – The names of the nodes or variables to be excluded from the simulation. By default, no nodes or variables are skipped. (default:
()
)
- Returns:
The model instance itself after updating its state with the simulated values.
- Raises:
AttributeError – If the value of the
Dist.at
node of a distribution node cannot be set.
Notes
The simulation is based on the shapes of the current values of the
Dist.at
nodes of the distribution nodes. If theDist.at
node of a distribution node is aVarValue
node, the value of its input is updated.