Sensitivity analysis in Python

# __author__ = 'Bayes Server'
# __version__= '0.2'

import jpype    # pip install jpype1    (version 1.2.1 or later)
import jpype.imports
from jpype.types import *

classpath = "lib/bayesserver-10.8.jar"  # TODO download the Bayes Server Java API, and adjust the path

# Launch the JVM
jpype.startJVM(classpath=[classpath])

# import the Java modules
from com.bayesserver import *
from com.bayesserver.inference import *
from com.bayesserver.analysis import *

# Uncomment the following line and change the license key, if you are using a licensed version
# License.validate("xxx")

# TODO download network from the Bayes Server User Interface (or Bayes Server Online)
# and adjust the following path
network_path = 'networks/Asia.bayes'

network = Network()
network.load(network_path)

variables = network.getVariables()

visit_to_asia = variables.get('Visit to Asia', True)
has_lung_cancer = variables.get('Has Lung Cancer', True)
tuberculosis_or_cancer = variables.get('Tuberculosis or Cancer', True)
smoker = variables.get('Smoker', True)
has_tuberculosis = variables.get('Has Tuberculosis', True)
dyspnea = variables.get('Dyspnea', True)
xray_result = variables.get('XRay Result', True)
has_bronchitis = variables.get('Has Bronchitis', True)

xRayResultAbnormal = xray_result.getStates().get('Abnormal', True)
smokerFalse = smoker.getStates().get('False', True)
hasLungCancerFalse = has_lung_cancer.getStates().get('False', True)

evidence = DefaultEvidence(network)

# TODO set any evidence here if you need to...

sensitivity = SensitivityToParameters(network, RelevanceTreeInferenceFactory())

parameter = ParameterReference(has_lung_cancer.getNode(), [smokerFalse, hasLungCancerFalse])

oneWay = sensitivity.oneWay(
    evidence,
    xRayResultAbnormal,
    parameter
)

print('Parameter value = {}'.format(oneWay.getParameterValue()))
print('Sensitivity value = {}'.format(oneWay.getSensitivityValue()))
print('P(Abnormal | e) = {}'.format(oneWay.getProbabilityHypothesisGivenEvidence()))
print('Alpha = {}'.format(oneWay.getAlpha()))
print('Beta = {}'.format(oneWay.getBeta()))
print('Delta = {}'.format(oneWay.getDelta()))
print('Gamma = {}'.format(oneWay.getGamma()))

print('Eval(0.2) = {}'.format(oneWay.evaluate(0.2)))
print('Eval\'(0.2) = {}'.format(oneWay.evaluateDeriv(0.2)))


#   Expected output...

#   Parameter value = 0.99
#   Sensitivity value = -0.460164
#   P(Abnormal | e) = 0.11029004
#   Alpha = -0.460164
#   Beta = 0.5658524
#   Delta = 1
#   Gamma = 0
#   Eval(0.2) = 0.4738196
#   Eval'(0.2) = -0.460164