Model#

class liesel.model.Model(*nodes_and_vars, grow=True, copy=False, to_float32=True, validate_log_prob_decomposition=True)[source]#

Bases: object

A model with a static graph.

Tip

If you have an existing model and want to make changes to it, you can use the Model.pop_nodes_and_vars() method to release the nodes and variables of the model. You can then make changes to them, for example i.e. changing the distribution of a variable or the inputs of a calculation. Afterwards, you initialize a new model with your changed variables. If you simply want to change the value of a variable, it is not necessary to call pop_nodes_and_vars(), you can simply override the Var.value attribute. Remeber to call Model.update() afterwards.

Parameters:

nodes_and_vars (Node | Var | Iterable[Node | Var]) – The nodes and variables to include in the model.
grow (bool, default: True) – Whether a GraphBuilder should be used to grow the model (finding the recursive inputs of the nodes and variables), and to add the model nodes.
copy (bool, default: False) – Whether the nodes and variables should be copied upon initialization.
to_float32 (bool, default: True) – Whether to convert the dtype of the values of the added nodes from float64 to float32. Only takes effect if grow=True.

See also

Var.new_obs: Initializes a strong variable that holds observed data.
Var.new_param: Initializes a strong variable that acts as a model parameter.
Var.new_calc: Initializes a weak variable that is a function of other variables.
Var.new_value: Initializes a strong variable without a distribution.
GraphBuilder: A graph builder, which can be used to set up and manipulate a model if you need more control.

Examples

Here, we set up a basic model based on three variables:

>>> a = lsl.Var.new_value(1.0, name="a")
>>> b = lsl.Var.new_value(2.0, name="b")
>>> c = lsl.Var.new_calc(lambda x, y: x + y, a, b, name="c")

We now build a model:

>>> model = lsl.Model([c])
>>> model
Model(9 nodes, 3 vars)

Methods

`copy`()	Returns a new model filled with deep copies of all model nodes and variables.
`copy_nodes_and_vars`()	Returns an unfrozen deep copy of the model nodes and variables.
`copy_vars`()	Returns an unfrozen deep copy of the model variables.
`diagnose`([verbose])	Provides a dataframe with diagnostic information about the model.
`extract_position`(position_keys[, model_state])	Extracts a position from a model state.
`groups`()	Collects the groups from all nodes and variables.
`node_parental_subgraph`(*of)	Returns a subgraph that consists of the input nodes and their parent nodes.
`parental_submodel`(*of)	Returns a new model that consists only of the given variables and nodes and their parent variables and nodes.
`plot`([show, save_path, width, height, prog, ...])	Plots the variables of this model.
`plot_nodes`([show, save_path, width, height, ...])	Plots the nodes of this model.
`plot_vars`([show, save_path, width, height, ...])	Plots the variables of this model.
`pop_nodes_and_vars`()	Pops the nodes and variables out of this model.
`pop_vars`()	Pops the variables out of this model.
`predict`(samples[, predict, newdata])	Returns a dictionary of predictions.
`sample`(shape, seed[, posterior_samples, ...])	Draws samples from the model.
`set_seed`(seed)	Splits and sets the seed / PRNG key.
`simulate`(seed[, skip])	Updates the model state simulating from the probability distributions in the model using a provided random seed, optionally skipping specified nodes.
`update`(*names)	Updates the target nodes and their recursive inputs if they are outdated.
`update_state`(position[, model_state, inplace])	Updates and returns a model state given a position.
`var_parental_subgraph`(*of)	Returns a subgraph that consists of the input variables and their parent variables.

Attributes

`auto_update`	Whether to update the model automatically if the value of a node is modified.
`log_lik`	The log-likelihood of the model.
`log_prior`	The log-prior of the model.
`log_prob`	The (unnormalized) log-probability / log-posterior of the model.
`model_nodes`	Dictionary of nodes with the `"_model"` prefix in their name.
`node_graph`	The directed graph of the model nodes.
`nodes`	A mapping of the model nodes with their names as keys.
`observed`	A mapping of the observed model variables with their names as keys.
`parameters`	A mapping of the model parameters with their names as keys.
`state`	The state of the model as a dict of node names and states.
`var_graph`	The directed graph of the model variables.
`vars`	A mapping of the model variables with their names as keys.

Model

Contents

Model#