def _buildKrigingAlgo(self, inputSample, outputSample):
"""
Build the functional chaos algorithm without running it.
"""
if self._basis is None:
# create linear basis only for the defect parameter (1st parameter),
# constant otherwise
input = ['x' + str(i) for i in range(self._dim)]
functions = []
# constant
functions.append(ot.NumericalMathFunction(input, ['y'], ['1']))
# linear for the first parameter only
functions.append(ot.NumericalMathFunction(input, ['y'],
[input[0]]))
self._basis = ot.Basis(functions)
if self._covarianceModel is None:
# anisotropic squared exponential covariance model
covColl = ot.CovarianceModelCollection(self._dim)
for i in range(self._dim):
if LooseVersion(ot.__version__) == '1.6':
covColl[i] = ot.SquaredExponential(1, 1.)
elif LooseVersion(ot.__version__) > '1.6':
covColl[i] = ot.SquaredExponential([1], [1.])
self._covarianceModel = ot.ProductCovarianceModel(covColl)
if LooseVersion(ot.__version__) == "1.9":
algoKriging = ot.KrigingAlgorithm(inputSample, outputSample,
self._covarianceModel,
self._basis)
else:
algoKriging = ot.KrigingAlgorithm(inputSample, outputSample,
self._basis,
self._covarianceModel, True)
algoKriging.run()
return algoKriging
# np.testing.assert_almost_equal(detectionSize1[1], 4.634627604344363, decimal=5)
# def test_1_Q2_90():
# np.testing.assert_almost_equal(POD1.getQ2(), 0.99993575194237017, decimal=4)
# Test kriging with censored data without Box Cox
np.random.seed(0)
ot.RandomGenerator.SetSeed(0)
ot.RandomGenerator.SetState(ot.RandomGeneratorState(ot.Indices([0] * 768), 0))
POD2 = otpod.KrigingPOD(inputSample,
signals,
detection,
noiseThres,
saturationThres,
boxCox=False)
if ot.__version__ == '1.6':
covColl = ot.CovarianceModelCollection(4)
scale = [5.03148, 13.9442, 20, 20]
for i in range(4):
c = ot.SquaredExponential(1, scale[i])
c.setAmplitude([15.1697])
covColl[i] = c
covarianceModel = ot.ProductCovarianceModel(covColl)
elif ot.__version__ > '1.6':
covarianceModel = ot.SquaredExponential([5.03148, 13.9442, 20, 20],
[15.1697])
POD2.setCovarianceModel(covarianceModel)
POD2.setInitialStartSize(0)
POD2.setSamplingSize(100)
POD2.setSimulationSize(100)
POD2.run()
detectionSize2 = POD2.computeDetectionSize(0.6, 0.95)
distX = ot.ComposedDistribution([X1, X2, X3])
# Get a sample of it
size = 100
X = distX.getSample(size)
# The Ishigami model
modelIshigami = ot.SymbolicFunction(
["X1", "X2", "X3"], ["sin(X1) + 5.0 * (sin(X2))^2 + 0.1 * X3^4 * sin(X1)"])
# Apply model: Y = m(X)
Y = modelIshigami(X)
# We define the covariance models for the HSIC indices.
# For the input, we consider a SquaredExponential covariance model.
covarianceModelCollection = ot.CovarianceModelCollection()
# Input sample
for i in range(3):
Xi = X.getMarginal(i)
Cov = ot.SquaredExponential(1)
Cov.setScale(Xi.computeStandardDeviation())
covarianceModelCollection.add(Cov)
# Output sample with squared exponential covariance
Cov2 = ot.SquaredExponential(1)
Cov2.setScale(Y.computeStandardDeviation())
covarianceModelCollection.add(Cov2)
# We choose an estimator type :
# - unbiased: HSICUStat;
distX = ot.ComposedDistribution([X1, X2, X3])
# Get a sample of it
size = 100
X = distX.getSample(size)
# The Ishigami model
modelIshigami = ot.SymbolicFunction(
["X1", "X2", "X3"], ["sin(X1) + 5.0 * (sin(X2))^2 + 0.1 * X3^4 * sin(X1)"])
# Apply model: Y = m(X)
Y = modelIshigami(X)
# We define the covariance models for the HSIC indices.
# For the input, we consider a SquaredExponential covariance model.
covarianceList = ot.CovarianceModelCollection()
# Input sample
for i in range(3):
Xi = X.getMarginal(i)
Cov = ot.SquaredExponential(1)
Cov.setScale(Xi.computeStandardDeviation())
covarianceList.add(Cov)
# Output sample with squared exponential covariance
Cov2 = ot.SquaredExponential(1)
Cov2.setScale(Y.computeStandardDeviation())
covarianceList.add(Cov2)
# We choose an estimator type :
# - unbiased: HSICUStat (not available here!!);
