Fix/message passing typo and docs (#1760)

* fix: typo BeliefPropagationWithMessageParsing to BeliefPropagationWithMessagePassing

* fix: add Belief Propagation with Message Passing to docs

* fix: typos where BeliefPropagationWithMessagePassing is called

* fix: len of ===

docs/exact_infer/base.rst

@@ -3,6 +3,7 @@ Exact Inference
 
 - :doc:`ve`
 - :doc:`bp`
+- :doc:`bp_wmp`
 - :doc:`causal`
 - :doc:`mplp`
 - :doc:`dbn_infer`
@@ -13,6 +14,7 @@ Exact Inference
 
    ve.rst
    bp.rst
+   bp_wmp.rst
    causal.rst
    mplp.rst
    dbn_infer.rst

docs/exact_infer/bp_wmp.rst

@@ -0,0 +1,5 @@
+Belief Propagation with Message Passing
+========================================
+
+.. autoclass:: pgmpy.inference.ExactInference.BeliefPropagationWithMessagePassing
+   :members:

pgmpy/inference/ExactInference.py

@@ -1238,11 +1238,11 @@ def map_query(
         return map_query_results
 
 
-class BeliefPropagationWithMessageParsing(Inference):
+class BeliefPropagationWithMessagePassing(Inference):
     """
     Class for performing efficient inference using Belief Propagation method on factor graphs with no loops.
 
-    The message-parsing algorithm recursively parses the factor graph to propagate the
+    The message-passing algorithm recursively parses the factor graph to propagate the
     model's beliefs to infer the posterior distribution of the queried variable. The recursion
     stops when reaching an observed variable or a unobserved root/leaf variable.
 
@@ -1269,7 +1269,7 @@ def __init__(self, model: FactorGraph, check_model=True):
 
     class _RecursiveMessageSchedulingQuery(object):
         """
-        Private class used in `BeliefPropagationWithMessageParsing.query()` to efficiently
+        Private class used in `BeliefPropagationWithMessagePassing.query()` to efficiently
         manage the message scheduling across the different queried variables, in a recursive way.
 
         Parameters

pgmpy/inference/__init__.py

@@ -2,7 +2,7 @@
 from .CausalInference import CausalInference
 from .ExactInference import BeliefPropagation
 from .ExactInference import VariableElimination
-from .ExactInference import BeliefPropagationWithMessageParsing
+from .ExactInference import BeliefPropagationWithMessagePassing
 from .ApproxInference import ApproxInference
 from .dbn_inference import DBNInference
 from .mplp import Mplp
@@ -12,7 +12,7 @@
     "VariableElimination",
     "DBNInference",
     "BeliefPropagation",
-    "BeliefPropagationWithMessageParsing",
+    "BeliefPropagationWithMessagePassing",
     "BayesianModelSampling",
     "CausalInference",
     "ApproxInference",

pgmpy/tests/test_inference/test_ExactInference.py

@@ -6,7 +6,7 @@
 
 from pgmpy.factors.discrete import DiscreteFactor, TabularCPD
 from pgmpy.inference import BeliefPropagation, VariableElimination
-from pgmpy.inference.ExactInference import BeliefPropagationWithMessageParsing
+from pgmpy.inference.ExactInference import BeliefPropagationWithMessagePassing
 from pgmpy.models import BayesianNetwork, FactorGraph, JunctionTree, MarkovNetwork
 
 
@@ -1113,7 +1113,7 @@ def tearDown(self):
         del self.bayesian_model
 
 
-class TestBeliefPropagationWithMessageParsing(unittest.TestCase):
+class TestBeliefPropagationWithMessagePassing(unittest.TestCase):
     def setUp(self):
         self.factor_graph = FactorGraph()
         self.factor_graph.add_nodes_from(["A", "B", "C", "D"])
@@ -1141,7 +1141,7 @@ def setUp(self):
             ]
         )
 
-        self.belief_propagation = BeliefPropagationWithMessageParsing(self.factor_graph)
+        self.belief_propagation = BeliefPropagationWithMessagePassing(self.factor_graph)
 
     def test_query_single_variable(self):
         res = self.belief_propagation.query(["C"])