Inference

class pgmpy.inference.Inference(model)

Bases: object

Base class for all inference algorithms.

Converts DiscreteBayesianNetwork and DiscreteMarkovNetwork to a uniform representation so that inference algorithms can be applied. Also, it checks if all the associated CPDs / Factors are consistent with the model.

Initialize inference for a model.

Parameters:

model: pgmpy.models.DiscreteBayesianNetwork or pgmpy.models.DiscreteMarkovNetwork

model for which to initialize the inference object.

Examples

>>> from pgmpy.inference import Inference
>>> from pgmpy.models import DiscreteBayesianNetwork
>>> from pgmpy.factors.discrete import TabularCPD
>>> student = DiscreteBayesianNetwork([("diff", "grade"), ("intel", "grade")])
>>> diff_cpd = TabularCPD("diff", 2, [[0.2], [0.8]])
>>> intel_cpd = TabularCPD("intel", 2, [[0.3], [0.7]])
>>> grade_cpd = TabularCPD(
...     "grade",
...     3,
...     [[0.1, 0.1, 0.1, 0.1], [0.1, 0.1, 0.1, 0.1], [0.8, 0.8, 0.8, 0.8]],
...     evidence=["diff", "intel"],
...     evidence_card=[2, 2],
... )
>>> student.add_cpds(diff_cpd, intel_cpd, grade_cpd)
>>> model = Inference(student)

>>> from pgmpy.models import DiscreteMarkovNetwork
>>> from pgmpy.factors.discrete import DiscreteFactor
>>> import numpy as np
>>> student = DiscreteMarkovNetwork(
...     [
...         ("Alice", "Bob"),
...         ("Bob", "Charles"),
...         ("Charles", "Debbie"),
...         ("Debbie", "Alice"),
...     ]
... )
>>> factor_a_b = DiscreteFactor(
...     ["Alice", "Bob"], cardinality=[2, 2], values=np.random.rand(4)
... )
>>> factor_b_c = DiscreteFactor(
...     ["Bob", "Charles"], cardinality=[2, 2], values=np.random.rand(4)
... )
>>> factor_c_d = DiscreteFactor(
...     ["Charles", "Debbie"], cardinality=[2, 2], values=np.random.rand(4)
... )
>>> factor_d_a = DiscreteFactor(
...     ["Debbie", "Alice"], cardinality=[2, 2], values=np.random.rand(4)
... )
>>> student.add_factors(factor_a_b, factor_b_c, factor_c_d, factor_d_a)
>>> model = Inference(student)