inferlo.GraphModel

class inferlo.GraphModel

Abstract class representing any graphical model.

__init__(num_variables: int, domain: Domain)

Parameters:

• num_variables – Number of variables in the model.

• domain – Default domain of each variable.

Methods

__init__(num_variables, domain)
add_factor(factor)	Adds a factor to the model.
draw_factor_graph(ax)	Draws the factor graph.
evaluate(x)	Returns value of non-normalized pdf in point.
get_factor_graph()	Builds factor graph for the model.
get_factors()	Returns all factors.
get_max_domain_size()	Returns the biggest domain size over all variables.
get_symbolic_variables()	Prepares variables for usage in expressions.
get_variable(idx)	Returns variable by its index.
get_variables()	Returns all variables.
infer([algorithm])	Performs inference.
max_likelihood([algorithm])	Finds the most probable state.
max_likelihood_bruteforce()	Evaluates most likely state in a very inefficient way.
part_func_bruteforce()	Evaluates partition function in very inefficient way.
sample(num_samples[, algorithm])	Generates samples.

abstractmethod add_factor(factor: Factor)

Adds a factor to the model.

draw_factor_graph(ax)

Draws the factor graph.

evaluate(x: ndarray) → float

Returns value of non-normalized pdf in point.

In other words, just substitutes values into factors and multiplies them.

get_factor_graph() → Tuple[Graph, Dict[int, str]]

Builds factor graph for the model.

Factor graph is a bipartite graph with variables in one part and factors in other graph. Edge denotes that factor depends on variable.

abstractmethod get_factors() → Iterable[Factor]

Returns all factors.

get_max_domain_size()

Returns the biggest domain size over all variables.

get_symbolic_variables() → List[FunctionFactor]

Prepares variables for usage in expressions.

Returns lists of trivial FunctionFactor s, each of them representing a factor on one variable with identity function. They can be used in mathematical expressions, which will result in another FunctionFactor.

get_variable(idx: int) → Variable

Returns variable by its index.

get_variables() → List[Variable]

Returns all variables.

abstractmethod infer(algorithm='auto', **kwargs)

Performs inference.

abstractmethod max_likelihood(algorithm='auto', **kwargs) → ndarray

Finds the most probable state.

max_likelihood_bruteforce()

Evaluates most likely state in a very inefficient way.

part_func_bruteforce()

Evaluates partition function in very inefficient way.

sample(num_samples: int, algorithm='auto', **kwargs) → ndarray

Generates samples.