Model creation and inspection#

Model(*args, **kwargs)

Encapsulates the variables and likelihood factors of a model.


Return the given model or, if none was supplied, try to find one in the context stack.


Deterministic(name, var[, model, dims])

Create a named deterministic variable.

Potential(name, var[, model, dims])

Add an arbitrary term to the model log-probability.

set_data(new_data[, model, coords])

Sets the value of one or more data container variables.

Point(*args[, filter_model_vars])

Build a point.

compile_fn(outs, *[, inputs, mode, ...])

Compiles an PyTensor function

Graph visualization#

model_to_networkx([model, var_names, ...])

Produce a networkx Digraph from a PyMC model.

model_to_graphviz([model, var_names, ...])

Produce a graphviz Digraph from a PyMC model.