#!/usr/bin/env python3
from itertools import chain
import numpy as np
import networkx as nx
[docs]class NoisyOrModel(nx.DiGraph):
"""
Base class for Noisy-Or models.
This is an implementation of generalized Noisy-Or models and
is not limited to Boolean variables and also any arbitrary
function can be used instead of the boolean OR function.
Reference: http://xenon.stanford.edu/~srinivas/research/6-UAI93-Srinivas-Generalization-of-Noisy-Or.pdf
"""
def __init__(self, variables, cardinality, inhibitor_probability):
# TODO: Accept values of each state so that it could be
# put into F to compute the final state values of the output
"""
Init method for NoisyOrModel.
Parameters
----------
variables: list, tuple, dict (array like)
array containing names of the variables.
cardinality: list, tuple, dict (array like)
array containing integers representing the cardinality
of the variables.
inhibitor_probability: list, tuple, dict (array_like)
array containing the inhibitor probabilities of each variable.
Examples
--------
>>> from pgmpy.models import NoisyOrModel
>>> model = NoisyOrModel(['x1', 'x2', 'x3'], [2, 3, 2], [[0.6, 0.4],
... [0.2, 0.4, 0.7],
... [0.1, 0.4]])
"""
self.variables = np.array([])
self.cardinality = np.array([], dtype=int)
self.inhibitor_probability = []
self.add_variables(variables, cardinality, inhibitor_probability)
[docs] def add_variables(self, variables, cardinality, inhibitor_probability):
"""
Adds variables to the NoisyOrModel.
Parameters
----------
variables: list, tuple, dict (array like)
array containing names of the variables that are to be added.
cardinality: list, tuple, dict (array like)
array containing integers representing the cardinality
of the variables.
inhibitor_probability: list, tuple, dict (array_like)
array containing the inhibitor probabilities corresponding to each variable.
Examples
--------
>>> from pgmpy.models import NoisyOrModel
>>> model = NoisyOrModel(['x1', 'x2', 'x3'], [2, 3, 2], [[0.6, 0.4],
... [0.2, 0.4, 0.7],
... [0.1, 0., 4]])
>>> model.add_variables(['x4'], [3], [0.1, 0.4, 0.2])
"""
if len(variables) == 1:
if not isinstance(inhibitor_probability[0], (list, tuple)):
inhibitor_probability = [inhibitor_probability]
if len(variables) != len(cardinality):
raise ValueError("Size of variables and cardinality should be same")
elif any(
cardinal != len(prob_array)
for prob_array, cardinal in zip(inhibitor_probability, cardinality)
) or len(cardinality) != len(inhibitor_probability):
raise ValueError(
"Size of variables and inhibitor_probability should be same"
)
elif not all(
0 <= item <= 1 for item in chain.from_iterable(inhibitor_probability)
):
raise ValueError(
"Probability values should be between 0 and 1(both inclusive)."
)
else:
self.variables = np.concatenate((self.variables, variables))
self.cardinality = np.concatenate((self.cardinality, cardinality))
self.inhibitor_probability.extend(inhibitor_probability)
[docs] def del_variables(self, variables):
"""
Deletes variables from the NoisyOrModel.
Parameters
----------
variables: list, tuple, dict (array like)
list of variables to be deleted.
Examples
--------
>>> from pgmpy.models import NoisyOrModel
>>> model = NoisyOrModel(['x1', 'x2', 'x3'], [2, 3, 2], [[0.6, 0.4],
... [0.2, 0.4, 0.7],
... [0.1, 0., 4]])
>>> model.del_variables(['x1'])
"""
variables = [variables] if isinstance(variables, str) else set(variables)
indices = [
index
for index, variable in enumerate(self.variables)
if variable in variables
]
self.variables = np.delete(self.variables, indices, 0)
self.cardinality = np.delete(self.cardinality, indices, 0)
self.inhibitor_probability = [
prob_array
for index, prob_array in enumerate(self.inhibitor_probability)
if index not in indices
]
#
# def out_prob(self, func):
# """
# Compute the conditional probability of output variable
# given all other variables [P(X|U)] where X is the output
# variable and U is the set of input variables.
#
# Parameters
# ----------
# func: function
# The deterministic function which maps input to the
# output.
#
# Returns
# -------
# List of tuples. Each tuple is of the form (state, probability).
# """
# states = []
# from itertools import product
# for u in product([(values(var)) for var in self.variables]):
# for state in product([(values(var) for var in self.variables)]):
